1 2004-11-04 Hans-Peter Nilsson <hp@axis.com>
3 * configure.in (crisv32): Recognize. AC_DEFINE_UNQUOTED
4 DEFAULT_CRIS_ARCH. Handle crisv32-*-linux-gnu* like
5 cris-*-linux-gnu* and crisv32-*-* like cris-*-*.
6 * configure: Regenerate.
7 * config/tc-cris.c (enum cris_archs): New.
8 (cris_mach, cris_arch_from_string, s_cris_arch, get_sup_reg)
9 (cris_insn_ver_valid_for_arch): New functions.
10 (DEFAULT_CRIS_ARCH): New macro, default to cris_any_v0_v10.
11 (cris_arch): New variable.
12 (md_pseudo_table): New pseudo .arch.
13 (err_for_dangerous_mul_placement): Initialize according to
15 (STATE_COND_BRANCH): Renamed from STATE_CONDITIONAL_BRANCH.
17 (STATE_COND_BRANCH_V32, STATE_COND_BRANCH_COMMON)
18 (STATE_ABS_BRANCH_V32, STATE_LAPC, BRANCH_BF_V32, BRANCH_BB_V32)
19 (BRANCH_WF_V32, BRANCH_WB_V32): New.
20 (BRANCH_BF, BRANCH_BB, BRANCH_WF, BRANCH_WB): Don't undef after
21 use in md_cris_relax_table.
22 (md_cris_relax_table): Add entries for STATE_COND_BRANCH_V32,
23 STATE_COND_BRANCH_COMMON, STATE_ABS_BRANCH_V32, STATE_LAPC.
24 Update and improve head comment.
25 (OPTION_PIC): Define in terms of previous option, OPTION_US.
26 (OPTION_MULBUG_ABORT_ON, OPTION_MULBUG_ABORT_OFF): Similar.
28 (md_longopts): New option --march=...
29 (cris_any_v0_v10_long_jump_size, crisv32_long_jump_size): New
31 (md_long_jump_size): Initialize in terms of DEFAULT_CRIS_ARCH.
32 (HANDLE_RELAXABLE): New macro.
33 (md_estimate_size_before_relax): Use HANDLE_RELAXABLE for common
34 cases. Check for weak symbols and assume not relaxable. Handle
35 STATE_COND_BRANCH_V32, STATE_COND_BRANCH_COMMON,
36 STATE_ABS_BRANCH_V32, STATE_LAPC. Use new variable symbolP, not
38 (md_convert_frag): Handle STATE_COND_BRANCH_V32,
39 STATE_COND_BRANCH_COMMON, STATE_ABS_BRANCH_V32, STATE_LAPC.
40 (cris_create_short_jump): Adjust for CRISv32.
41 (cris_relax_frag): Handle new states.
42 (md_create_long_jump): Ditto. Emit error for common_v10_v32.
43 (md_begin): Define symbols "..asm.arch.cris.v32",
44 "..asm.arch.cris.v10", "..asm.arch.cris.common_v10_v32" and
45 "..asm.arch.cris.any_v0_v10". Use cris_insn_ver_valid_for_arch
46 when entering opcode table entry points.
47 (md_assemble): Adjust branch handling for CRISv32. Handle LAPC
48 relaxation. In fix_new_exp call for main insn, pass 1 for pcrel
49 parameter for 8, 16 and 32-bit pc-relative insns and LAPC.
50 (cris_process_instruction): Initialize out_insnp->insn_type to
51 CRIS_INSN_NONE, not CRIS_INSN_NORMAL.
52 <case ']', '[', 'A', 'd', 'Q', 'N', 'n', 'Y', 'U', 'u', 'T'>: New
54 <case 'm'>: Check that modified_char == '.'.
55 <invalid operands>: Consume the rest of the line.
56 When operands don't match, skip over subsequent insns with
57 non-matching version specifier but same mnemonic.
58 <immediate constant, case SIZE_SPEC_REG>: Immediate operands for
59 special registers in CRISv32 are always 32 bit long.
60 <immediate constant, case SIZE_FIELD_SIGNED, SIZE_FIELD_UNSIGNED>:
62 (get_gen_reg): Only recognize "PC" when followed by "+]" for v32
63 and compatible. Recognize "ACR" for v32, unless followed by "+".
64 (get_spec_reg): Consider cris_arch when looking up register.
65 (get_autoinc_prefix_or_indir_op): Don't recognize assignment for
67 (get_3op_or_dip_prefix_op): Check for ']' after seeing '[rN+'.
68 (cris_get_expression): Restore input_line_pointer if failing "early".
69 (get_flags): Consider cris_arch and recognize flags accordingly.
70 (branch_disp): Adjust for CRISv32.
71 (gen_cond_branch_32): Similar. Emit error for common_v10_v32.
72 (cris_number_to_imm): Use as_bad_where, not as_bad. Remove
73 related FIXME. Don't insist on BFD_RELOC_32_PCREL fixup to be
74 resolved. Don't enter zeros in object file for
76 <case BFD_RELOC_CRIS_LAPCQ_OFFSET, BFD_RELOC_CRIS_SIGNED_16>
77 <case BFD_RELOC_CRIS_SIGNED_8>: New case.
78 (md_parse_option): Break out "return 1".
79 <OPTION_ARCH> New case.
80 (tc_gen_reloc): <case BFD_RELOC_CRIS_LAPCQ_OFFSET>
81 <case BFD_RELOC_CRIS_SIGNED_16, BFD_RELOC_CRIS_SIGNED_8>
82 <case BFD_RELOC_CRIS_UNSIGNED_8, BFD_RELOC_CRIS_UNSIGNED_16>
83 <case BFD_RELOC_32_PCREL>: New cases.
84 Addends for non-zero fx_pcrel are too in fx_offset.
85 (md_show_usage): Show --march=<arch>.
86 (md_apply_fix3): Adjust val for BFD_RELOC_CRIS_LAPCQ_OFFSET.
87 (md_pcrel_from): BFD_RELOC_CRIS_LAPCQ_OFFSET is PC-relative too.
88 (s_syntax) <struct syntaxes>: Properly constify member operand.
89 * config/tc-cris.h (TARGET_MACH): Define.
91 * doc/as.texinfo (Overview) <CRIS>: Add --march=...
92 * doc/c-cris.texi (CRIS-Symbols): New node for built-in symbols.
93 (CRIS-Opts): Document --march=...
94 (CRIS-Pseudos): Document .arch.
96 2004-11-04 Jan Beulich <jbeulich@novell.com>
98 * config/tc-i386.c (set_intel_syntax): Allow % in symbol names when
99 intel syntax and no register prefix, allow $ in symbol names when
101 (set_16bit_gcc_code_flag): Replace literal 'l' by LONG_MNEM_SUFFIX.
102 (intel_float_operand): Add fourth return value indicating math control
103 operations. Make classification more precise.
104 (md_assemble): Complain if memory operand of mov[sz]x has no size
106 (parse_insn): Translate word operands to floating point instructions
107 operating on integers as well as control instructions to short ones
108 as expected by AT&T syntax. Translate 'd' suffix to short one only for
109 floating point instructions operating on non-integer operands.
110 (match_template): Remove fldcw special case. Adjust q-suffix handling
111 to permit it on fild/fistp/fisttp in AT&T mode.
112 (process_suffix): Don't guess DefaultSize insns' suffix from
113 stackop_size for certain floating point control instructions. Guess
114 suffix for branch and [ls][gi]dt based on flag_code. Split error
115 messages for Intel and AT&T syntax, and make the condition more strict
116 for the former. Adjust suppressing of generation of operand size
118 (intel parser): Allow the full set of MASM operators. Add FWORD, TBYTE,
119 OWORD, and XMMWORD operand size specifiers (TBYTE replaces XWORD). Add
121 * config/tc-i386.h (BYTE_PTR WORD_PTR DWORD_PTR QWORD_PTR XWORD_PTR
122 SHORT OFFSET_FLAT FLAT NONE_FOUND): Remove unused defines.
124 2004-11-03 Hans-Peter Nilsson <hp@axis.com>
126 * symbols.c (colon) [!WORKING_DOT_WORD]: Don't declare
127 md_short_jump_size, md_long_jump_size.
128 * write.c [!WORKING_DOT_WORD]: Ditto.
129 * tc.h [!WORKING_DOT_WORD]: Declare them here. Drop const
131 * config/tc-cris.h (md_short_jump_size, md_long_jump_size): Don't
133 * config/tc-cris.c (md_short_jump_size, md_long_jump_size): Drop
134 const qualifier in these definitions.
135 * config/tc-i370.c, config/tc-m68k.c, config/tc-pdp11.c,
136 config/tc-s390.c, config/tc-tahoe.c, config/tc-vax.c: Ditto.
138 2004-11-02 Nick Clifton <nickc@redhat.com>
140 * dwarf2dbg.c (dwarf2_finish): Check for the existence of a file
141 table before deciding to produce a .debug_line section to match up
142 with a user provided .debug_info section.
144 2004-10-28 Tomer Levi <Tomer.Levi@nsc.com>
146 * config/tc-crx.c (getreg_image): Bug fix, a return value was
147 mistakenly omitted from CRX_C_REGTYPE and CRX_CS_REGTYPE cases.
149 2004-10-27 Tomer Levi <Tomer.Levi@nsc.com>
151 * config/tc-crx.c: Remove global variable 'post_inc_mode'.
152 (get_flags): New function.
153 (get_number_of_bits): Edit comments, update numeric values to supported
155 (process_label_constant): Don't support the colon format (SYMBOL:[s|m|l]).
156 (set_cons_rparams): Support argument type 'arg_rbase'.
157 (get_operandtype): Bug fix in 'rbase' operand type parsing.
158 (handle_LoadStor): Bug fix, first handle post-increment mode.
159 (getreg_image): Remove redundant code, update according to latest CRX spec.
160 (print_constant): Bug fix relate to 3-word instructions.
161 (assemble_insn): Bug fix, when matching instructions, verify also
162 instruction type (not only mnemonic).
163 Add various error checking.
164 (preprocess_reglist): Support HI/LO and user registers.
166 2004-10-25 David Mosberger-Tang <davidm@hpl.hp.com>
168 * config/tc-ia64.c (fixup_unw_records): Don't let the "t" value
169 in an epilogue directive go negative.
171 2004-10-25 H.J. Lu <hongjiu.lu@intel.com>
174 * config/tc-ia64.c (emit_one_bundle): Decrement md.num_slots_in_use
175 after reporting template error during manual bundling. Reported
176 by Michael Dupont, michaelx.dupont@intel.com.
178 2004-10-25 Daniel Jacobowitz <dan@debian.org>
180 * Makefile.am: Run dep-am.
181 * aclocal.m4: Regenerate with automake 1.9.2.
182 * Makefile.in: Regenerate with automake 1.9.2.
183 * doc/Makefile.in: Likewise.
185 * config/tc-arm.c: Include "dw2gencfi.h".
186 (tc_arm_regname_to_dw2regnum, tc_arm_frame_initial_instructions):
188 * config/tc-arm.h (TARGET_USE_CFIPOP, DWARF2_DEFAULT_RETURN_COLUMN)
189 (DWARF2_CIE_DATA_ALIGNMENT, tc_regname_to_dw2regnum)
190 (tc_cfi_frame_initial_instructions): Define.
191 (tc_arm_regname_to_dw2regnum, tc_arm_frame_initial_instructions):
194 2004-10-21 Tomer Levi <Tomer.Levi@nsc.com>
196 * config/tc-crx.c (assemble_insn): Check unsigned immediate
198 Update coprocessor id to be unsigned immediate.
200 2004-10-18 Aaron W. LaFramboise <aaron98wiridge9@aaronwl.com>
202 * config/tc-i386.c (O_secrel): Delete.
203 (tc_pe_dwarf2_emit_offset): New function.
204 * config/tc-i386.h (O_secrel): Define as O_md1.
205 (TC_DWARF2_EMIT_OFFSET): Define.
207 2004-10-18 Nick Clifton <nickc@redhat.com>
209 * config/tc-xstormy16.c (xstormy16_cons_fix_new): Accept and
210 ignore @fptr() directives for 4-byte fixups.
212 2004-10-15 Alan Modra <amodra@bigpond.net.au>
214 * Makefile.am: Run "make dep-am"
215 * Makefile.in: Regenerate.
217 2004-10-14 Bob Wilson <bob.wilson@acm.org>
219 * doc/c-xtensa.texi (Xtensa Options, Absolute Literals Directive):
220 Remove comments about placement of literal pools.
221 (Literal Directive): Update description of literal placement.
222 (Literal Prefix Directive): Remove statement that this does not apply
223 to absolute-mode literals. Describe new section naming scheme.
225 2004-10-12 Bob Wilson <bob.wilson@acm.org>
227 * config/tc-xtensa.c (xtensa_frequency_pseudo): Use set_subseg_freq.
228 (is_entry_opcode, is_movi_opcode, is_the_loop_opcode, is_jx_opcode,
229 is_windowed_return_opcode): Delete.
230 (xtensa_frob_label): Use get_subseg_target_freq.
231 (md_assemble): Inline call to is_entry_opcode.
232 (xtensa_handle_align): Inline call to get_frag_is_literal.
233 (relaxation_requirements): Inline call to is_jx_opcode.
234 (emit_single_op): Inline call to is_movi_opcode.
235 (xg_assemble_vliw_tokens): Inline calls to get_frag_is_insn,
236 get_frag_is_no_transform, is_entry_opcode, and
237 set_frag_is_specific_opcode. Use get_subseg_total_freq.
238 (xtensa_fix_a0_b_retw_frags, xtensa_fix_b_j_loop_end_frags,
239 xtensa_fix_close_loop_end_frags, relax_frag_immed, convert_frag_immed):
240 Inline calls to get_frag_is_no_transform.
241 (next_instrs_are_b_retw): Inline call to is_windowed_return_opcode.
242 (xtensa_fix_short_loop_frags): Inline calls to is_the_loop_opcode and
243 get_frag_is_no_transform.
244 (convert_frag_immed_finish_loop): Inline calls to get_expression_value
245 and set_frag_is_no_transform.
246 (get_expression_value): Delete.
247 (subseg_map struct): Rename cur_total_freq to total_freq. Rename
248 cur_target_freq to target_freq.
249 (get_subseg_info): Split out code to create a new map entry into ...
250 (add_subseg_info): ... this new function.
251 (get_last_insn_flags): Check if get_subseg_info succeeded.
252 (set_last_insn_flags): Call add_subseg_info if needed.
253 (get_subseg_total_freq, get_subseg_target_freq, set_subseg_freq): New.
254 (xtensa_reorder_segments): Compute last_sec while counting sections.
255 Remove call to get_last_sec.
256 (get_last_sec): Delete.
257 (cache_literal_section): Inline call to retrieve_literal_seg and its
258 callees, seg_present and add_seg_list.
259 (retrieve_literal_seg, seg_present, add_seg_list): Delete.
260 (get_frag_is_insn, get_frag_is_no_transform,
261 set_frag_is_specific_opcode, set_frag_is_no_transform): Delete.
262 * config/tc-xtensa.h (MAX_SLOTS): Reduce from 31 to 15.
264 2004-10-12 Bob Wilson <bob.wilson@acm.org>
266 * config/tc-xtensa.c: Use ISO C90 formatting.
267 * config/tc-xtensa.h: Likewise.
268 * config/xtensa-istack.h: Likewise.
269 * config/xtensa-relax.c: Likewise.
270 * config/xtensa-relax.h: Likewise.
272 2004-10-12 Paul Brook <paul@codesourcery.com>
274 * config/tc-arm.c (md_begin): Change EF_ARM_EABI_VER3 to
277 * doc/c-arm.texi: Document that we actually support -meabi=4, not
280 2004-10-08 Bob Wilson <bob.wilson@acm.org>
282 * doc/as.texinfo (VTableEntry, VTableInherit): Add "directive" to index
284 (Acknowledgements): Use "GAS" instead of AS variable.
286 2004-10-08 Daniel Jacobowitz <dan@debian.org>
288 * config/tc-i386.c: Include "elf/x86-64.h".
289 (i386_elf_section_type): New function.
290 * config/tc-i386.h (md_elf_section_type): Define.
291 (i386_elf_section_type): New prototype.
293 2004-10-08 Linus Nielsen Feltzing <linus@haxx.se>
295 * config/m68k-parse.h (enum m68k_register): New control register,
297 * config/tc-m68k.c: Correct control register set for MCF5249.
299 2004-10-07 Bob Wilson <bob.wilson@acm.org>
300 Sterling Augustine <sterling@tensilica.com>
302 * config/tc-xtensa.c (absolute_literals_supported): New global flag.
303 (UNREACHABLE_MAX_WIDTH): Define.
304 (XTENSA_FETCH_WIDTH): Delete.
305 (cur_vinsn, xtensa_fetch_width, xt_saved_debug_type, past_xtensa_end,
306 prefer_const16, prefer_l32r): New global variables.
307 (LIT4_SECTION_NAME): Define.
308 (lit4_state struct): Add lit4_seg_name and lit4_seg fields.
309 (XTENSA_PROP_*, GET_XTENSA_PROP_*, SET_XTENSA_PROP_*): Define.
310 (frag_flags struct): New.
311 (xtensa_block_info struct): Move from tc-xtensa.h. Add flags field.
312 (subseg_map struct): Add cur_total_freq and cur_target_freq fields.
313 (bitfield, bit_is_set, set_bit, clear_bit): Define.
314 (MAX_FORMATS): Define.
315 (op_placement_info struct, op_placement_table): New.
316 (O_pltrel, O_hi16, O_lo16): Define.
317 (directiveE enum): Rename directive_generics to directive_transform.
318 Delete directive_relax. Add directive_schedule,
319 directive_absolute_literals, and directive_last_directive.
320 (directive_info): Rename "generics" to "transform". Delete "relax".
321 Add "schedule" and "absolute-literals".
322 (directive_state): Adjust entries to match changes in directive_info.
323 (xtensa_relax_statesE, RELAX_IMMED_MAXSTEPS): Move to tc-xtensa.h.
324 (xtensa_const16_opcode, xtensa_movi_opcode, xtensa_movi_n_opcode,
325 xtensa_l32r_opcode, xtensa_nop_opcode, xtensa_rsr_lcount_opcode): New.
326 (xtensa_j_opcode, xtensa_rsr_opcode): Delete.
327 (align_only_targets, software_a0_b_retw_interlock,
328 software_avoid_b_j_loop_end, maybe_has_b_j_loop_end,
329 software_avoid_short_loop, software_avoid_close_loop_end,
330 software_avoid_all_short_loops, specific_opcode): Delete.
331 (warn_unaligned_branch_targets): New.
332 (workaround_a0_b_retw, workaround_b_j_loop_end, workaround_short_loop,
333 workaround_close_loop_end, workaround_all_short_loops): Default FALSE.
334 (option_[no_]link_relax, option_[no_]transform,
335 option_[no_]absolute_literals, option_warn_unaligned_targets,
336 option_prefer_l32r, option_prefer_const16, option_target_hardware):
338 (option_[no_]align_only_targets, option_literal_section_name,
339 option_text_section_name, option_data_section_name,
340 option_bss_section_name, option_eb, option_el): Delete.
341 (md_longopts): Add entries for: [no-]transform, [no-]absolute-literals,
342 warn-unaligned-targets, prefer-l32r, prefer-const16, [no-]link-relax,
343 and target-hardware. Delete entries for [no-]target-align-only,
344 literal-section-name, text-section-name, data-section-name, and
346 (md_parse_option): Handle new options and remove old ones. Accept but
347 ignore [no-]density options. Warn for [no-]generics and [no-]relax
348 and treat them as [no-]transform.
349 (md_show_usage): Add new options and remove old ones.
350 (xtensa_setup_hw_workarounds): New.
351 (md_pseudo_table): Change "word" entry to use xtensa_elf_cons. Add
352 "long", "short", "loc" and "frequency" entries.
353 (use_generics): Rename to ...
354 (use_transform): ... this function. Add past_xtensa_end check.
355 (use_longcalls): Add past_xtensa_end check.
356 (code_density_available, can_relax): Delete.
357 (do_align_targets): New.
358 (get_directive): Accept dashes in directive names. Warn about
359 [no-]generics and [no-]relax directives and treat them as
361 (xtensa_begin_directive): Call md_flush_pending_output only for some
362 directives. Check for directives inside instruction bundles. Warn
363 about deprecated ".begin literal" usage. Warn and ignore [no-]density
364 directives. Handle new directives. Check generating_literals flag
366 (xtensa_end_directive): Check for directives inside instruction
367 bundles. Warn and ignore [no-]density directives. Handle new
368 directives. Call xtensa_set_frag_assembly_state.
369 (xtensa_loc_directive_seen, xtensa_dwarf2_directive_loc,
370 xtensa_dwarf2_emit_insn): New.
371 (xtensa_literal_position): Call md_flush_pending_output. Do not check
372 use_literal_section flag.
373 (xtensa_literal_pseudo): Call md_flush_pending_output. Handle absolute
374 literals. Use xtensa_elf_cons to parse the expression.
375 (xtensa_literal_prefix): Do not check use_literal_section. Support
376 ".lit4" sections for absolute literals. Change prefix convention to
377 replace ".text" (or ".t" in a linkonce section). No need to call
379 (xtensa_frequency_pseudo, xtensa_elf_cons, xtensa_elf_suffix): New.
380 (expression_end): Handle closing braces and colons.
381 (PLT_SUFFIX, plt_suffix): Delete.
382 (expression_maybe_register): Use new xtensa-isa.h functions. Use
383 xtensa_elf_suffix instead of checking for plt suffix, and handle O_lo16
384 and O_hi16 expressions as well.
385 (tokenize_arguments): Handle closing braces and colons.
386 (parse_arguments): Use new xtensa-isa.h functions. Handle "invisible"
387 operands and paired register syntax.
388 (get_invisible_operands): New.
389 (xg_translate_sysreg_op): Handle new Xtensa LX RSR/WSR/XSR syntax. Use
390 new xtensa-isa.h functions.
391 (xtensa_translate_old_userreg_ops, xtensa_translate_zero_immed): New.
392 (xg_translate_idioms): Check if inside bundle. Use use_transform.
393 Handle new Xtensa LX RSR/WSR/XSR syntax. Remove code to widen density
394 instructions. Use xtensa_translate_zero_immed.
395 (operand_is_immed, operand_is_pcrel_label): Delete.
396 (get_relaxable_immed): Use new xtensa-isa.h functions.
397 (get_opcode_from_buf): Add slot parameter. Use new xtensa-isa.h
399 (xtensa_print_insn_table, print_vliw_insn): New.
400 (is_direct_call_opcode): Use new xtensa-isa.h functions.
401 (is_call_opcode, is_loop_opcode, is_conditional_branch_opcode,
402 is_branch_or_jump_opcode): Delete.
403 (is_movi_opcode, decode_reloc, encode_reloc, encode_alt_reloc): New.
404 (opnum_to_reloc, reloc_to_opnum): Delete.
405 (xtensa_insnbuf_set_operand, xtensa_insnbuf_get_operand): Use new
406 xtensa-isa.h functions. Operate on one slot of an instruction.
407 (xtensa_insnbuf_set_immediate_field, is_negatable_branch,
408 xg_get_insn_size): Delete.
409 (xg_get_build_instr_size): Use xg_get_single_size.
410 (xg_is_narrow_insn, xg_is_single_relaxable_insn): Update calls to
411 xg_build_widen_table. Use xg_get_single_size.
412 (xg_get_max_narrow_insn_size): Delete.
413 (xg_get_max_insn_widen_size, xg_get_max_insn_widen_literal_size,
414 xg_is_relaxable_insn): Update calls to xg_build_widen_table. Use
416 (xg_build_to_insn): Record the loc field. Handle OP_OPERAND_HI16U and
417 OP_OPERAND_LOW16U. Check xg_valid_literal_expression.
418 (xg_expand_to_stack, xg_expand_narrow): Update calls to
419 xg_build_widen_table. Use xg_get_single_size.
420 (xg_immeds_fit): Use new xtensa-isa.h functions. Update call to
422 (xg_symbolic_immeds_fit): Likewise. Also handle O_lo16 and O_hi16, and
423 treat weak symbols conservatively.
424 (xg_check_operand): Use new xtensa-isa.h functions.
425 (is_dnrange): Delete.
426 (xg_assembly_relax): Inline previous calls to tinsn_copy.
427 (xg_finish_frag): Specify separate relax states for the frag and slot0.
428 (is_branch_jmp_to_next, xg_add_branch_and_loop_targets): Use new
429 xtensa-isa.h functions.
430 (xg_instruction_matches_option_term, xg_instruction_matches_or_options,
431 xg_instruction_matches_options): New.
432 (xg_instruction_matches_rule): Handle O_register expressions. Call
433 xg_instruction_matches_options.
434 (transition_rule_cmp): New.
435 (xg_instruction_match): Update call to xg_build_simplify_table.
436 (xg_build_token_insn): Record loc fields.
437 (xg_simplify_insn): Check is_specific_opcode field and
438 density_supported flag.
439 (xg_expand_assembly_insn): Skip checking code_density_available. Use
440 new xtensa-isa.h functions. Call use_transform instead of can_relax.
441 (xg_assemble_literal): Add error handling for O_big. Call
442 record_alignment. Handle O_pltrel.
443 (xg_valid_literal_expression): New.
444 (xg_assemble_literal_space): Add slot parameter. Remove call to
445 set_expr_symbol_offset. Add call to record_alignment. Update call to
447 (xg_emit_insn): Delete.
448 (xg_emit_insn_to_buf): Add format parameter. Update calls to
449 xg_add_opcode_fix and xtensa_insnbuf_to_chars.
450 (xg_add_opcode_fix): Change opcode parameter to tinsn and add format
451 and slot parameters. Handle new "alternate" relocations for absolute
452 literals and CONST16 instructions. Check for bad uses of O_lo16 and
453 O_hi16. Use new xtensa-isa.h functions.
454 (xg_assemble_tokens): Delete.
455 (is_register_writer): Use new xtensa-isa.h functions.
456 (is_bad_loopend_opcode): Check for xtensa_rsr_lcount_opcode instead of
457 old-style RSR from LCOUNT.
458 (next_frag_opcode): Delete.
459 (next_frag_opcode_is_loop, next_frag_format_size, frag_format_size,
460 update_next_frag_state): New.
461 (update_next_frag_nop_state): Delete.
462 (next_frag_pre_opcode_bytes): Use next_frag_opcode_is_loop.
463 (xtensa_mark_literal_pool_location): Check use_literal_section flag and
464 the state of the absolute-literals directive. Add calls to
465 record_alignment and xtensa_set_frag_assembly_state. Call
466 xtensa_switch_to_non_abs_literal_fragment instead of
467 xtensa_switch_to_literal_fragment.
469 (assemble_nop): Use build_nop. Update call to xtensa_insnbuf_to_chars.
470 (get_expanded_loop_offset): Change check for undefined opcode to an
472 (xtensa_set_frag_assembly_state, relaxable_section,
473 xtensa_find_unmarked_state_frags, xtensa_find_unaligned_branch_targets,
474 xtensa_find_unaligned_loops, xg_apply_tentative_value): New.
475 (md_begin): Update call to xtensa_isa_init. Initialize linkrelax to 1.
476 Set lit4_seg_name. Call xg_init_vinsn. Initialize new global opcodes.
477 Call init_op_placement_info_table and xtensa_set_frag_assembly_state.
478 (xtensa_init_fix_data): New.
479 (xtensa_frob_label): Reset label symbol to the current frag. Check
480 do_align_targets and generating_literals flag. Propagate frequency
481 info to new alignment frag. Call xtensa_set_frag_assembly_state.
482 (xtensa_unrecognized_line): New.
483 (xtensa_flush_pending_output): Check if inside a bundle. Add a call
484 to xtensa_set_frag_assembly_state.
485 (error_reset_cur_vinsn): New.
486 (md_assemble): Remove check for literal frag. Remove call to
487 istack_init. Call use_transform instead of use_generics. Parse
488 explicit instruction format specifiers. Move code for
489 a0_b_retw_interlock workaround to xg_assemble_vliw_tokens. Call
490 error_reset_cur_vinsn on errors. Add call to get_invisible_operands.
491 Add dwarf2_where call. Remote automatic alignment for ENTRY
492 instructions. Move call to xtensa_clear_insn_labels to the end.
493 Rearrange to handle bundles.
494 (xtensa_cons_fix_new): Delete.
495 (xtensa_handle_align): New.
496 (xtensa_frag_init): Call xtensa_set_frag_assembly_state. Remove
497 assignment to is_no_density field.
498 (md_pcrel_from): Use new xtensa-isa.h functions. Use decode_reloc
499 instead of reloc_to_opnum. Handle "alternate" relocations.
500 (xtensa_force_relocation, xtensa_check_inside_bundle,
501 xtensa_elf_section_change_hook): New.
502 (xtensa_symbol_new_hook): Delete.
503 (xtensa_fix_adjustable): Check for difference of symbols with an
504 offset. Check for external and weak symbols.
505 (md_apply_fix3): Remove cases for XTENSA_OP{0,1,2} relocs.
506 (md_estimate_size_before_relax): Return expansion for the first slot.
507 (tc_gen_reloc): Handle difference of symbols by producing
508 XTENSA_DIFF{8,16,32} relocs and by writing the value of the difference
509 into the output. Handle new XTENSA_SLOT*_OP relocs by storing the
510 tentative values into the output when linkrelax is set.
511 (XTENSA_PROP_SEC_NAME): Define.
512 (xtensa_post_relax_hook): Call xtensa_find_unmarked_state_frags.
513 Create literal tables only if using literal sections. Create new
514 property tables instead of old instruction tables. Check for unaligned
515 branch targets and loops.
516 (finish_vinsn, find_vinsn_conflicts, check_t1_t2_reads_and_writes,
517 new_resource_table, clear_resource_table, resize_resource_table,
518 resources_available, reserve_resources, release_resources,
519 opcode_funcUnit_use_unit, opcode_funcUnit_use_stage,
520 resources_conflict, xg_find_narrowest_format, relaxation_requirements,
521 bundle_single_op, emit_single_op, xg_assemble_vliw_tokens): New.
522 (xtensa_end): Call xtensa_flush_pending_output. Set past_xtensa_end
523 flag. Update checks for workaround options. Call
524 xtensa_mark_narrow_branches and xtensa_mark_zcl_first_insns.
525 (xtensa_cleanup_align_frags): Add special case for branch targets.
526 Check for and mark unreachable frags.
527 (xtensa_fix_target_frags): Remove use of align_only_targets flag.
528 Use RELAX_LOOP_END_BYTES in special case for negatable branch at the
529 end of a zero-overhead loop body.
530 (frag_can_negate_branch): Handle instructions with multiple slots.
531 Use new xtensa-isa.h functions
532 (xtensa_mark_narrow_branches, is_narrow_branch_guaranteed_in_range,
533 xtensa_mark_zcl_first_insns): New.
534 (xtensa_fix_a0_b_retw_frags, xtensa_fix_b_j_loop_end_frags): Error if
535 transformations are disabled.
536 (next_instrs_are_b_retw): Use new xtensa-isa.h functions. Handle
537 multislot instructions.
538 (xtensa_fix_close_loop_end_frags, xtensa_fix_short_loop_frags):
539 Likewise. Also error if transformations are disabled.
540 (unrelaxed_frag_max_size): New.
541 (unrelaxed_frag_min_insn_count, unrelax_frag_has_b_j): Use new
542 xtensa-isa.h functions.
543 (xtensa_sanity_check, is_empty_loop, is_local_forward_loop): Use
544 xtensa_opcode_is_loop instead of is_loop_opcode.
545 (get_text_align_power): Replace as_fatal with assertion.
546 (get_text_align_fill_size): Iterate instead of using modulus when
548 (get_noop_aligned_address): Assert that this is for a machine-dependent
549 RELAX_ALIGN_NEXT_OPCODE frag. Use next_frag_opcode_is_loop,
550 xg_get_single_size, and frag_format_size.
551 (get_widen_aligned_address): Rename to ...
552 (get_aligned_diff): ... this function. Add max_diff parameter.
553 Remove handling of rs_align/rs_align_code frags. Use
554 next_frag_format_size, get_text_align_power, get_text_align_fill_size,
555 next_frag_opcode_is_loop, and xg_get_single_size. Compute max_diff
556 and pass it back to caller.
557 (xtensa_relax_frag): Use relax_frag_loop_align. Add code for new
558 RELAX_SLOTS, RELAX_MAYBE_UNREACHABLE, RELAX_MAYBE_DESIRE_ALIGN,
559 RELAX_FILL_NOP, and RELAX_UNREACHABLE frag types. Check relax_seen.
560 (relax_frag_text_align): Rename to ...
561 (relax_frag_loop_align): ... this function. Assume loops can only be
562 in the first slot of an instruction.
563 (relax_frag_add_nop): Use assemble_nop instead of constructing an OR
564 instruction. Remove call to frag_wane.
565 (relax_frag_narrow): Rename to ...
566 (relax_frag_for_align): ... this function. Extend to handle
567 RELAX_FILL_NOP and RELAX_UNREACHABLE, as well as RELAX_SLOTS with
568 RELAX_NARROW for the first slot.
569 (find_address_of_next_align_frag, bytes_to_stretch): New.
570 (future_alignment_required): Use find_address_of_next_align_frag and
571 bytes_to_stretch. Look ahead to subsequent frags to make smarter
573 (relax_frag_immed): Add format, slot, and estimate_only parameters.
574 Check if transformations are enabled for b_j_loop_end workaround.
575 Use new xtensa-isa.h functions and handle multislot instructions.
576 Update call to xg_assembly_relax.
577 (md_convert_frag): Handle new RELAX_SLOTS, RELAX_UNREACHABLE,
578 RELAX_MAYBE_UNREACHABLE, RELAX_MAYBE_DESIRE_ALIGN, and RELAX_FILL_NOP
580 (convert_frag_narrow): Add segP, format and slot parameters. Call
581 convert_frag_immed for branch instructions. Adjust calls to
582 tinsn_from_chars, tinsn_immed_from_frag, and xg_emit_insn_to_buf. Use
583 xg_get_single_size and xg_get_single_format.
584 (convert_frag_fill_nop): New.
585 (convert_frag_immed): Add format and slot parameters. Handle multislot
586 instructions and use new xtensa-isa.h functions. Update calls to
587 tinsn_immed_from_frag and xg_assembly_relax. Check if transformations
588 enabled for b_j_loop_end workaround. Use build_nop instead of
589 assemble_nop. Check is_specific_opcode flag. Check for unreachable
590 frags. Use xg_get_single_size. Handle O_pltrel.
591 (fix_new_exp_in_seg): Remove check for old plt flag.
592 (convert_frag_immed_finish_loop): Update calls to tinsn_from_chars and
593 xtensa_insnbuf_to_chars. Call tinsn_immed_from_frag. Change check
594 for loop opcode to an assertion. Mark all frags up to the end of the
595 loop as not transformable.
596 (get_last_insn_flags, set_last_insn_flags): Use get_subseg_info.
597 (get_subseg_info): New.
598 (xtensa_move_literals): Call xtensa_set_frag_assembly_state. Add null
600 (xtensa_switch_to_literal_fragment): Rewrite to handle absolute
601 literals and use xtensa_switch_to_non_abs_literal_fragment otherwise.
602 (xtensa_switch_to_non_abs_literal_fragment): New.
603 (cache_literal_section): Add is_code parameter and pass it through to
604 retrieve_literal_seg.
605 (retrieve_literal_seg): Add is_code parameter and use it to set the
606 flags on the literal section. Handle case where head parameter is 0.
607 (get_frag_is_no_transform, set_frag_is_specific_opcode,
608 set_frag_is_no_transform): New.
609 (xtensa_create_property_segments): Add end_property_function parameter
610 and pass it through to add_xt_block_frags. Call bfd_get_section_flags
611 and skip SEC_DEBUGGING and !SEC_ALLOC sections.
612 (xtensa_create_xproperty_segments, section_has_xproperty): New.
613 (add_xt_block_frags): Add end_property_function parameter and call it
614 if it is non-zero. Call xtensa_frag_flags_init.
615 (xtensa_frag_flags_is_empty, xtensa_frag_flags_init,
616 get_frag_property_flags, frag_flags_to_number,
617 xtensa_frag_flags_combinable, xt_block_aligned_size,
618 xtensa_xt_block_combine, add_xt_prop_frags,
619 init_op_placement_info_table, opcode_fits_format_slot,
620 xg_get_single_size, xg_get_single_format): New.
621 (istack_push): Inline call to tinsn_copy.
622 (tinsn_copy): Delete.
623 (tinsn_has_invalid_symbolic_operands): Handle O_hi16 and O_lo16 and
624 CONST16 opcodes. Handle O_big, O_illegal, and O_absent.
625 (tinsn_has_complex_operands): Handle O_hi16 and O_lo16.
626 (tinsn_to_insnbuf): Use xg_get_single_format and new xtensa-isa.h
627 functions. Handle invisible operands.
628 (tinsn_to_slotbuf): New.
629 (tinsn_check_arguments): Use new xtensa-isa.h functions.
630 (tinsn_from_chars): Add slot parameter. Rewrite using xg_init_vinsn,
631 vinsn_from_chars, and xg_free_vinsn.
632 (tinsn_from_insnbuf): New.
633 (tinsn_immed_from_frag): Add slot parameter and handle multislot
634 instructions. Handle symbol differences.
635 (get_num_stack_text_bytes): Use xg_get_single_size.
636 (xg_init_vinsn, xg_clear_vinsn, vinsn_has_specific_opcodes,
637 xg_free_vinsn, vinsn_to_insnbuf, vinsn_from_chars, expr_is_register,
638 get_expr_register, set_expr_symbol_offset_diff): New.
639 * config/tc-xtensa.h (MAX_SLOTS): Define.
640 (xtensa_relax_statesE): Move from tc-xtensa.c. Add
641 RELAX_CHECK_ALIGN_NEXT_OPCODE, RELAX_MAYBE_DESIRE_ALIGN, RELAX_SLOTS,
642 RELAX_FILL_NOP, RELAX_UNREACHABLE, RELAX_MAYBE_UNREACHABLE, and
644 (RELAX_IMMED_MAXSTEPS): Move from tc-xtensa.c.
645 (xtensa_frag_type struct): Add is_assembly_state_set,
646 use_absolute_literals, relax_seen, is_unreachable, is_specific_opcode,
647 is_align, is_text_align, alignment, and is_first_loop_insn fields.
648 Replace is_generics and is_relax fields by is_no_transform field.
649 Delete is_text and is_longcalls fields. Change text_expansion and
650 literal_expansion to arrays of MAX_SLOTS entries. Add arrays of
651 per-slot information: literal_frags, slot_subtypes, slot_symbols,
652 slot_sub_symbols, and slot_offsets. Add fr_prev field.
653 (xtensa_fix_data struct): New.
654 (xtensa_symfield_type struct): Delete plt field.
655 (xtensa_block_info struct): Move definition to tc-xtensa.h. Add
656 forward declaration here.
657 (xt_section_type enum): Delete xt_insn_sec. Add xt_prop_sec.
658 (XTENSA_SECTION_RENAME): Undefine.
659 (TC_FIX_TYPE, TC_INIT_FIX_DATA, TC_FORCE_RELOCATION, NO_PSEUDO_DOT,
660 tc_unrecognized_line, md_do_align, md_elf_section_change_hook,
661 HANDLE_ALIGN, TC_LINKRELAX_FIXUP, SUB_SEGMENT_ALIGN): Define.
662 (TC_CONS_FIX_NEW, tc_symbol_new_hook): Delete.
663 (unit_num_copies_func, opcode_num_units_func,
664 opcode_funcUnit_use_unit_func, opcode_funcUnit_use_stage_func): New.
665 (resource_table struct): New.
666 * config/xtensa-istack.h (MAX_INSN_ARGS): Increase from 6 to 10.
667 (TInsn struct): Add keep_wide, loc, fixup, record_fix, subtype,
668 literal_space, symbol, sub_symbol, offset, and literal_frag fields.
669 (tinsn_copy): Delete prototype.
670 (vliw_insn struct): New.
671 * config/xtensa-relax.c (insn_pattern_struct): Add options field.
672 (widen_spec_list): Add option conditions for density and boolean
673 instructions. Add expansions using CONST16 and conditions for using
674 CONST16 vs. L32R. Use new Xtensa LX RSR/WSR syntax. Add entries for
676 (simplify_spec_list): Add option conditions for density instructions.
677 Add entry for NOP instruction.
678 (append_transition): Add cmp function pointer parameter and use it to
679 insert the new entry in order.
680 (operand_function_LOW16U, operand_function_HI16U): New.
681 (xg_has_userdef_op_fn, xg_apply_userdef_op_fn): Handle
682 OP_OPERAND_LOW16U and OP_OPERAND_HI16U.
683 (enter_opname, split_string): Use xstrdup instead of strdup.
684 (init_insn_pattern): Initialize new options field.
685 (clear_req_or_option_list, clear_req_option_list,
686 clone_req_or_option_list, clone_req_option_list, parse_option_cond):
688 (parse_insn_pattern): Parse option conditions.
689 (transition_applies): New.
690 (build_transition): Use new xtensa-isa.h functions. Fix incorrectly
691 swapped last arguments in calls to append_constant_value_condition.
692 Call clone_req_option_list. Add warning about invalid opcode.
693 Handle LOW16U and HI16U function names.
694 (build_transition_table): Add cmp parameter and use it in calls to
695 append_transition. Use new xtensa-isa.h functions. Check
696 transition_applies before adding entries.
697 (xg_build_widen_table, xg_build_simplify_table): Add cmp parameter and
698 pass it through to build_transition_table.
699 * config/xtensa-relax.h (ReqOrOptionList, ReqOrOption, ReqOptionList,
700 ReqOption, transition_cmp_fn): New types.
701 (OpType enum): Add OP_OPERAND_LOW16U and OP_OPERAND_HI16U.
702 (transition_rule struct): Add options field.
703 * doc/as.texinfo (Overview): Update Xtensa options.
704 * doc/c-xtensa.texi (Xtensa Options): Delete --[no-]density,
705 --[no-]relax, and --[no-]generics options. Update descriptions of
706 --text-section-literals and --[no-]longcalls. Add
707 --[no-]absolute-literals and --[no-]transform.
708 (Xtensa Syntax): Add description of syntax for FLIX instructions.
709 Remove use of "generic" and "specific" terminology for opcodes.
710 (Xtensa Registers): Generalize the syntax description to include
711 user-defined register files.
712 (Xtensa Automatic Alignment): Update.
713 (Xtensa Branch Relaxation): Mention limitation of unconditional jumps.
714 (Xtensa Call Relaxation): Linker can now remove most of the overhead.
715 (Xtensa Directives): Remove confusing rules about precedence.
716 (Density Directive, Relax Directive): Delete.
717 (Schedule Directive): New.
718 (Generics Directive): Rename to ...
719 (Transform Directive): ... this node.
720 (Literal Directive): Update for absolute literals. Missing
721 literal_position directive is now an error.
722 (Literal Position Directive): Update for absolute literals.
723 (Freeregs Directive): Delete.
724 (Absolute Literals Directive): New.
725 (Frame Directive): Minor editing.
726 * Makefile.am (DEPTC_xtensa_elf, DEPOBJ_xtensa_elf, DEP_xtensa_elf):
728 * Makefile.in: Regenerate.
730 2004-10-07 Richard Sandiford <rsandifo@redhat.com>
732 * config/tc-mips.c (append_insn): Use fix_new rather than fix_new_exp
733 to build the second and third fixups for a composite relocation.
734 (macro_read_relocs): New function.
735 (macro_build): Use it.
736 (s_cpsetup): Pass all three composite relocation codes to macro_build.
737 Simplify fragging code accordingly.
738 (s_gpdword): Use fix_new rather than fix_new_exp for the second part
739 of the composite relocation. Set fx_tcbit in both fixups.
741 2004-10-07 Richard Sandiford <rsandifo@redhat.com>
743 * config/tc-mips.c (append_insn): Set fx_tcbit for composite relocs.
744 (md_apply_fix3): Don't treat composite relocs as done.
746 2004-10-07 Jan Beulich <jbeulich@novell.com>
748 * macro.c (macro_expand_body): When ELF, use .LL rather than LL as
749 prefix for symbol names generated from the LOCAL macro directive.
751 * dw2gencfi.c (select_cie_for_fde): When separating CIE out from
752 FDE, treat a DW_CFA_remember_state as we do a DW_CFA_advance_loc.
754 2004-10-07 Tomer Levi <Tomer.Levi@nsc.com>
756 * config/tc-crx.c (preprocess_reglist): Handle Co-processor
758 (md_assemble): Add error checking for Co-Processor instructions.
759 (get_cinv_parameters): Add 'b' option to invalidate the
762 2004-10-05 Paul Brook <paul@codesourcery.com>
764 * config/tc-arm.c (unwind): New variable.
765 (vfp_sp_encode_reg): New function.
766 (vfp_sp_reg_required_here): Use it.
767 (vfp_sp_reg_list, vfp_dp_reg_list): Remove.
768 (vfp_parse_reg_list): New function.
769 (s_arm_unwind_fnstart, s_arm_unwind_fnend, s_arm_unwind_cantunwind,
770 s_arm_unwind_personality, s_arm_unwind_personalityindex,
771 s_arm_unwind_handlerdata, s_arm_unwind_save, s_arm_unwind_movsp,
772 s_arm_unwind_pad, s_arm_unwind_setfp, s_arm_unwind_raw): New
774 (md_pseudo_table): Add them.
775 (do_vfp_reg2_from_sp2): Use vfp_parse_reg_list and vfp_sp_encode_reg.
776 (do_vfp_sp2_from_reg2, vfp_sp_ldstm, vfp_dp_ldstm): Ditto.
777 (set_section, add_unwind_adjustsp, flush_pending_unwind,
778 finish_unwind_opcodes, start_unwind_section, create_unwind_entry,
779 require_hashconst, add_unwind_opcode): New functions.
780 * doc/tc-arm.text: Document unwinding opcodes.
781 * NEWS: Mention the new feature.
783 2004-10-04 Eric Christopher <echristo@redhat.com>
785 * config/tc-mips.c (md_apply_fix3): Remove erroneous assert.
787 2004-10-01 H.J. Lu <hongjiu.lu@intel.com>
789 * config/tc-ppc.c (md_apply_fix3): Call S_SET_THREAD_LOCAL for
791 * config/tc-s390.c (md_apply_fix3): Likewise.
792 * config/tc-sparc.c (md_apply_fix3): Likewise.
794 2004-10-01 Paul Brook <paul@codesourcery.com>
796 * config/tc-arm.c (arm_elf_section_type): New function.
797 (arm_elf_change_section): Set section link for exidx sections.
798 * config/tc-arm.h (arm_elf_section_type): Add prototype.
799 (md_elf_section_type): Define.
801 2004-10-01 Bill Farmer <Bill@the-farmers.freeserve.co.uk>
803 * config/tc-pdp11.c (md_apply_fix3): Change to sign of the SOB
804 instruction's offset.
806 2004-10-01 Adam Nemet <anemet@lnxw.com>
808 * (TARGET_FORMAT): Remove LynxOS COFF definition.
810 2004-10-01 Ravi Ramaseshan <ravi.ramaseshan@codito.com>
812 * config/tc-arc.c (tc_gen_reloc): Don't assume fixP->fx_addsy is an
813 asymbol *, instead use symbol_get_bfdsym.
815 2004-09-30 Linus Nielsen Feltzing <linus@haxx.se>
817 * config/tc-m68k.c (select_control_regs): Add mcf5249.
819 2004-09-30 Paul Brook <paul@codesourcery.com>
821 * config/tc-arm.c (do_smi, do_nop): New functions.
822 (insns): Add ARMv6ZK instructions.
823 (md_apply_fix3): Handle BFD_RELOC_ARM_SMI.
824 (tc_gen_reloc): Ditto.
825 (arm_cpus): Add mpcore and arm1176.
826 (arm_archs): Add armv6{k,z,zk}.
827 * doc/c-arm.texi: Document new cores and architectures.
829 2004-09-30 Nick Clifton <nickc@redhat.com>
831 * config/tc-arm.c: Use ISO C90 formatting.
833 2004-09-30 Vladimir Ivanov <vladitx@nucleusys.com>
835 * config/tc-arm.c (mav_reg_required_here): Allow REG_TYPE_CN
836 as alternative when REG_TYPE_MVF, REG_TYPE_MVD, REG_TYPE_MVFX or
837 REG_TYPE_MVDX is expected.
839 2004-09-29 Marc Bevand <m.bevand@gmail.com>
841 * doc/c-i386.texi (i386-Mnemonics): Fix typo.
843 2004-09-21 James E Wilson <wilson@specifixinc.com>
845 * config/tc-ia64.c (ENCODED_PSP_OFFSET): New.
846 (output_rp_psprel, output_pfs_psprel, output_preds_psprel,
847 output_spill_base, output_unat_psprel, output_lc_psprel,
848 output_fpsr_psprel, output_priunat_psprel, output_bsp_psprel,
849 output_bsprestore_psprel, output_rnat_psprel, output_spill_psprel,
850 output_spill_psprel_p): Use it.
852 2004-09-20 Tomer Levi <Tomer.Levi@nsc.com>
854 * config/tc-crx.c (handle_LoadStor): New function.
855 Handle load/stor unique instructions before parsing.
857 2004-09-17 Paul Brook <paul@codesourcery.com>
859 * config/tc-arm.c (s_arm_rel31): New funciton.
860 (md_pseudo_table): Add .rel31.
861 (md_apply_fix3): Handle BFD_RELOC_ARM_TARGET2,
862 BFD_RELOC_32_PCREL and BFD_RELOC_ARM_PREL31.
863 (tc_gen_reloc): Handle BFD_RELOC_ARM_PREL31 and BFD_RELOC_ARM_TARGET2.
864 (arm_fix_adjustable): Return 0 for BFD_RELOC_ARM_TARGET2.
865 (arm_parse_reloc): Add (target2).
867 2004-09-17 Alan Modra <amodra@bigpond.net.au>
869 * Makefile.am: Run "make dep-am".
870 * Makefile.in: Regenerate.
871 * aclocal.m4: Regenerate.
872 * configure: Regenerate.
873 * doc/Makefile.in: Regenerate.
874 * po/POTFILES.in: Regenerate.
875 * po/gas.pot: Regenerate.
877 2004-09-14 Hideki IWAMOTO <h-iwamoto@kit.hi-ho.ne.jp>
879 * config/tc-mmix.c [!LLONG_MIN]: Correct #elsif to #elif.
882 2004-09-13 Paul Brook <paul@codesourcery.com>
884 * config/tc-arm.c: Rename RELABS to TARGET1.
886 2004-09-13 Alan Modra <amodra@bigpond.net.au>
888 * messages.c (as_internal_value_out_of_range): Cast values passed
889 to as_bad_where or as_warn_where to proper type.
891 2004-09-11 Theodore A. Roth <troth@openavr.org>
893 * config/tc-avr.c: Add support for
894 atmega48, atmega88, atmega168, attiny13, attiny2313, at90can128.
896 2004-09-09 Alan Modra <amodra@bigpond.net.au>
898 * dw2gencfi.c (select_cie_for_fde): When separating CIE out
899 from FDE, treat a CFI_escape as we do a DW_CFA_advance_loc.
901 2004-09-08 Paul Brook <paul@codesourcery.com>
903 * config/obj-elf.c (obj_elf_section_type): Handle init_array,
904 fini_array and preinit_array section types.
905 * config/tc-ia64.c (ia64_elf_section_type): Remove init_array
907 * doc/as.texinfo: Document extra section types.
909 2004-09-02 Mark Mitchell <mark@codesourcery.com>
911 * Makefile.am (TARG_ENV_HFILES): Add te-symbian.h.
912 * Makefile.in: Regenerated.
913 * configure.in: Set em for arm*-*-symbianelf*.
914 * configure: Regenerated.
915 * config/tc-arm.c (elf32_arm_target_format): Use Symbian target
916 vectors when appropriate.
917 * config/te-symbian.h: New file.
919 2004-09-03 Tomer Levi <Tomer.Levi@nsc.com>
921 * config/tc-crx.c (gettrap): Exception vector can be case
923 (process_label_constant): Fix a 32-bit displacement bug in branch
925 (get_operandtype) : Bug fix, wrong operand was used.
926 (process_label_constant): Initialize relocation type to
929 2004-09-01 Richard Earnshaw < reanrsha@arm.com>
931 * tc-arm.c (arm_cpus, arm_fpus): Allow <cpu>-s as well as <cpu>s
932 for synthesizable cores.
934 * doc/c-arm.texi (ARM Options): Document canonical names of CPUs.
936 2004-08-25 Dmitry Diky <diwil@spec.ru>
938 * config/tc-msp430.c: Clean-up the code.
939 (md_relax_table): New relax table.
940 (mcu_types): Sort MCU types.
941 (md_pseudo_table): Add .profiler pseudo handler.
942 (pow2value): New function.
943 (msp430_profiler): New function.
944 (msp430_operands): Add new insns handlers.
945 (msp430_srcoperand): Add register operand handler, allow complex
947 (md_estimate_size_before_relax): Rewritten.
948 (md_convert_frag): Rewritten.
949 (msp430_relax_frag): New function.
950 * config/tc-msp430.h (md_relax_frag): define macro
951 * doc/c-msp430.texi: Update information.
953 2004-08-24 Nick Clifton <nickc@redhat.com>
955 * as.c (std_shortopts): Allow -g to take an optional argument.
956 (parse_args): Pass any switch starting with -g on to the backend
959 2004-08-18 Mark Mitchell <mark@codesourcery.com>
961 * configure.in (arm*-*-symbianelf*): New target.
962 (arm*-*-eabi*): Likewise.
963 * configure: Regenerated.
965 2004-08-18 Thiemo Seufer <seufer@csv.ica.uni-stuttgart.de>
966 * config/tc-mips.c (append_insn): Handle delay slots in branch likely
969 2004-08-18 Jakub Jelinek <jakub@redhat.com>
971 * config/tc-ia64.c (start_unwind_section): Add linkonce_empty
972 argument, don't do anything if current section is not
973 .gnu.linkonce.t.* and linkonce_empty is set.
974 (generate_unwind_image, dot_endp): Adjust callers, call
975 start_unwind_section (*, 1) if nothing will be put into the
978 2004-08-17 Nick Clifton <nickc@redhat.com>
980 * as.c (MD_DEBUG_FORMAT_SELECTOR): Provide default definition.
981 (show_usage): Add -g.
982 (std_longopts): Add --gen-debug. Alpha sort the table.
983 (parse_args): Print an error message if a switch is not handled.
984 Handle the -g switch, calling md_debug_format_selector() if
986 * NEWS: Mention new feature.
987 * doc/as.texinfo: Document new switch.
988 * doc/internals.texi: Document behaviour of md_parse_option.
990 * config/tc-arm.c (md_parse_option): Do not issue an error message
991 if the switch is not recognised.
992 * config/tc-m68k.c (md_parse_option): Likewise.
993 * config/tc-pdp11.c (md_parse_option): Likewise.
994 * config/tc-v850.c (md_parse_option): Likewise.
996 * as.h: Fix up formatting.
999 2004-08-16 Nick Clifton <nickc@redhat.com>
1001 * macro.c (macro_set_alternate): Use ISO C90 formatting.
1003 * configure.in: Sort architecture based tables alphabetically.
1004 * configure: Regenerate.
1006 2004-08-16 Alan Modra <amodra@bigpond.net.au>
1008 * config/tc-ppc.c (tc_ppc_regname_to_dw2regnum <regnames>): Replace
1009 { "cc", 68 }, with { "cr", 70 }.
1011 2004-08-13 Jan Beulich <jbeulich@novell.com>
1012 Nick Clifton <nickc@redhat.com>
1014 * as.c: Add and handle new --alternate command line option.
1015 * macro.c (macro_set_alternate): New.
1016 * macro.h (macro_set_alternate): Declare.
1017 * read.c: Add and handle new .altmacro and .noaltmacro directives.
1018 * doc/as.texinfo: Document new command line option and pseudo-ops
1019 as well as insert documentation originating from gasp about
1020 alternate macro syntax.
1021 * NEWS: Mention new command line option and pseudo-ops.
1023 2004-08-10 Mark Mitchell <mark@codesourcery.com>
1025 * expr.c (operand): Handle the "~", "-", and "!" operators applied
1028 2004-08-06 Paul Brook <paul@codesourcery.com>
1030 * config/tc-arm.c (md_apply_fix3, tc_gen_reloc, arm_parse_reloc):
1031 Handle new relocations.
1032 * include/elf/arm.h (elf_arm_reloc_type): Add new EABI relocations.
1034 2004-08-05 Bob Wilson <bob.wilson@acm.org>
1036 * write.c (relax_segment): Use was_address instead of address when
1037 setting fr_fix field for align frag due to backwards .org.
1039 2004-07-29 Alexandre Oliva <aoliva@redhat.com>
1041 Introduce SH2a support.
1042 2004-02-24 Corinna Vinschen <vinschen@redhat.com>
1043 * config/tc-sh.c (get_specific): Change arch_sh2a_up to
1045 2004-02-24 Corinna Vinschen <vinschen@redhat.com>
1046 * config/tc-sh.c (md_parse_option): Add sh2a-nofpu ISA handling.
1047 2004-02-20 Corinna Vinschen <vinschen@redhat.com>
1048 * config/tc-sh.c (sh_elf_final_processing): Move sh2a recognition
1049 to end of conditional expression.
1050 2004-02-20 Corinna Vinschen <vinschen@redhat.com>
1051 * config/tc-sh.c: Add sh2a-nofpu support.
1052 2003-12-29 DJ Delorie <dj@redhat.com>
1053 * tc-sh.c: Add sh2a support.
1054 (parse_reg): Add tbr.
1055 (parse_at): Support @@(disp,tbr).
1056 (get_specific): Support sh2a opcodes.
1057 (insert4): New, for 4 byte relocs.
1058 (build_Mytes): Support sh2a opcodes.
1059 (md_apply_fix3_Mytes): Support sh2a opcodes.
1060 2003-12-02 Michael Snyder <msnyder@redhat.com>
1061 * config/tc-sh.c (md_parse_option): Handle sh2a.
1062 (sh_elf_final_processing): Ditto.
1064 2004-07-27 Jason Thorpe <thorpej@wasabisystems.com>
1066 * config/tc-hppa.h (TARGET_FORMAT): Set to "elf32-hppa-netbsd"
1069 2004-07-27 Alan Modra <amodra@bigpond.net.au>
1071 * config/tc-ppc.c (ppc_frob_file_before_adjust): Warn if .toc too big.
1072 (ppc_arch): Expand comment.
1074 2004-07-27 Tomer Levi <Tomer.Levi@nsc.com>
1076 * config/tc-crx.c: Support evaluating the difference between two
1078 * config/tc-crx.h: Likewise.
1080 2004-07-26 H.J. Lu <hongjiu.lu@intel.com>
1082 * config/tc-ia64.c (start_unwind_section): Set the linked-to
1084 (ia64_elf_section_change_hook): Set the linked-to section for
1087 2004-07-26 Dmitry Diky <diwil@spec.ru>
1089 * config/tc-msp430.c: Add new subtargets: msp430x1610,
1090 msp430x1611, msp430x1612, msp430x415, msp430x417, msp430xG437,
1091 msp430xG438, msp430xG439.
1093 2004-07-25 Daniel Jacobowitz <dan@debian.org>
1095 * doc/as.texinfo (Section, PushSection): Correct documentation
1098 2004-07-21 Jan Beulich <jbeulich@novell.com>
1100 * config/tc-i386.c (optimize_imm): Adjust immediates to only those
1101 permissible for the selected instruction suffix.
1102 (match_template): Don't permit 64-bit general purpose operands in
1104 (finalize_imm): Permit 64-bit immediates.
1105 (build_modrm_byte): Don't treat 32-bit addressing in 64-bit mode
1106 specially except for the width of the used base and/or index
1107 registers. For 32-bit displacements, use sign-extended
1108 relocations only when using 64-bit addressing.
1109 Force zero displacement on rip-relative addressing when there is
1110 no other displacement.
1111 (i386_index_check): Don't treat 32-bit addressing in 64-bit mode
1112 specially except for the width of the used base and/or index
1114 (parse_register): Disallow Reg64 registers in 32-bit mode.
1116 * config/tc-i386.c: For DefaultSize instructions, don't guess a 'q'
1117 suffix if the instruction doesn't support it.
1119 2004-07-20 Maciej W. Rozycki <macro@linux-mips.org>
1121 * config/tc-mips.c (append_insn): Handle constant expressions with
1122 no associated relocation.
1123 (mips_ip): Cancel the expression after use for the Q format
1125 (parse_relocation): Return no relocation for unsupported
1127 (my_getSmallExpression): Return no relocation if no relocation
1130 2004-07-19 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
1132 * config/obj-som.c (adjust_stab_sections): Add prototype.
1133 (obj_som_compiler, obj_som_version, obj_som_copyright,
1134 adjust_stab_sections): Add ATTRIBUTE_UNUSED to unused arguments.
1135 * config/tc-hppa.c (update_subspace): Likewise.
1136 (is_defined_subspace): Amplify comment.
1137 * config/obj-som.h (som_frob_file): Add prototype.
1139 2004-07-19 Christopher Faylor <cgf@timesys.com>
1140 H.J. Lu <hongjiu.lu@intel.com>
1142 * subsegs.c (section_symbol): Don't create a new segment when
1143 existing segment is undefined.
1145 2004-07-16 Richard Earnshaw <rearnsha@arm.com>
1147 * config/tc-arm.c: Include include/opcode/arm.h.
1148 (ARM_EXT_*, ARM_ARCH_*, ARM_ANY, ARM_ALL, COPROC_ANY): Delete.
1149 (FPU_FPA_EXT_* FPU_VFP_EXT_*, FPU_ANY, FPU_NONE, FPU_MAVERICK): Delete.
1150 (FPU_ARCH_*): Delete.
1151 * Makefile.am: Update dependencies.
1152 * Makefile.in: Regenerate.
1154 2004-07-15 Nick Clifton <nickc@redhat.com>
1156 * configure.in: Accept armbe as a big-endian arm configuration.
1157 * configure: Regenerate.
1159 2004-07-13 Thomas Nystrom <thn@saeab.se>
1161 * config/tc-i386.c (T_SHIFTOP): New constant.
1162 (intel_e05_1): Handle '&', '|' and T_SHIFTOP.
1163 (intel_el1): Handle '~'.
1164 (intel_get_token): Handle '<>', '&', '|' and '~'.
1166 2004-07-13 Nick Clifton <nickc@redhat.com>
1168 (md_assemble): Remove spurious newline from end of as_bad error
1170 (intel_e05_1): Likewise.
1171 (intel_e11): Likewise.
1172 (intel_match_token): Likewise.
1174 2004-07-11 Andreas Schwab <schwab@suse.de>
1176 * config/tc-m68k.c: Convert to C90. Remove redundant
1177 declarations. Indentation fixup.
1178 [M68KCOFF]: Include "obj-coff.h" instead of declaring
1179 obj_coff_section ourselves.
1181 2004-07-09 James E Wilson <wilson@specifixinc.com>
1183 * config/tc-ia64.c (default_big_endian): New.
1184 (dot_byteorder, md_begin): Use it.
1185 (md_parse_option): Set it.
1187 2004-07-09 Nick Clifton <nickc@redhat.com>
1189 * configure.in: Change sh-sybmian-elf to sh-*-symbianelf.
1190 * configure: Regenerate.
1191 * NEWS: Change sh-sybmian-elf to sh-*-symbianelf.
1192 * config/tc-sh.c (sh_elf_final_processing): Use renamed version of
1193 sh_find_elf_flags if necessary.
1195 2004-07-08 Richard Sandiford <rsandifo@redhat.com>
1197 * config/tc-mips.c (mips_fix_adjustable): If the full addend is
1198 going to be split into more than one in-place addend, return 0
1199 for relocations against mergeable sections. Associate comments
1202 2004-07-07 Tomer Levi <Tomer.Levi@nsc.com>
1204 * Makefile.am (CPU_TYPES): Add crx.
1205 (TARGET_CPU_CFILES): Add config/tc-crx.c.
1206 (TARGET_CPU_HFILES): Add config/tc-crx.h.
1207 (DEPTC_crx_elf): New target.
1208 (DEPOBJ_crx_elf): Likewise.
1209 (DEP_crx_elf): Likewise.
1210 * Makefile.in: Regenerate.
1211 * configure.in: Add crx* target.
1212 * configure: Regenerate.
1213 * config/tc-crx.c: New file.
1214 * config/tc-crx.h: New file.
1215 * NEWS: Mention new target.
1217 2004-07-06 Nick Clifton <nickc@redhat.com>
1219 * config.in: Undefine TARGET_SYMBIAN by default.
1221 * configure: Regenerate. Add sh-symbian-elf target. If
1222 selected define TARGET_SYMBIAN.
1223 * config/tc-sh.h (TARGET_FORMAT): Select a Symbian target
1224 format if TARGET_SYMBIAN has been defined.
1226 * output-file.c (output_file_create): Report the target format
1227 chosen when bfd_openw reports that it is invalid.
1229 * config/obj-coff.c (coff_pseudo_table): Only define the weak
1230 pseudo for BFD based assemblers.
1232 2004-07-05 Andrew Stubbs <andrew.stubbs@superh.com>
1235 * config/tc-sh.c (md_assemble): Change isspace to ISSPACE.
1236 (md_parse_option): Remove redundant -isa testing.
1237 Make bfd_arch variable const.
1238 (md_show_usage): Make bfd_arch variable const.
1240 2004-07-03 James E Wilson <wilson@specifixinc.com>
1242 * config/tc-ia64.c (emit_one_bundle): Check and set insn_addr.
1243 * config/tc-ia64.h (md_frag_check): Define.
1245 2004-07-03 Aaron W. LaFramboise <aaron98wiridge9@aaronwl.com>
1247 * config/obj-coff.c (obj_coff_weak): New .weak syntax for PE weak
1249 * doc/as.texinfo (Weak): Document PE weak symbols.
1251 2004-07-03 Richard Sandiford <rsandifo@redhat.com>
1253 * config/tc-mips.c (HAVE_IN_PLACE_ADDENDS): New macro.
1254 (reloc_needs_lo_p): Only return true if HAVE_IN_PLACE_ADDENDS.
1255 (mips_frob_file): Rework so that only a single pass through the
1256 relocs is needed. Allow %lo()s to have higher offsets than their
1257 corresponding %hi()s or %got()s.
1259 2004-07-02 Nick Clifton <nickc@redhat.com>
1261 * config/tc-arm.c (md_apply_fix3:BFD_RELOC_ARM_IMMEDIATE): Do not
1262 allow values which have come from undefined symbols.
1263 Always consider this fixup to have been processed as a reloc
1264 cannot be generated for it.
1266 2004-07-02 Alan Modra <amodra@bigpond.net.au>
1268 * frags.h (struct frag): Add has_code and insn_addr fields.
1269 * write.c (cvt_frag_to_fill): Invoke md_frag_check.
1270 * config/tc-ppc.c (md_assemble): Check and set insn_addr.
1271 * config/tc-ppc.h (md_frag_check): Define.
1273 2004-06-28 Maciej W. Rozycki <macro@linux-mips.org>
1275 * doc/Makefile.am (info): Rename goal to...
1276 (info-local): ... this, to preserve implicit dependencies.
1277 * doc/Makefile.in: Regenerate with automake 1.8.5.
1279 2004-06-25 Kazuhiro Inaoka <inaoka.kazuhiro@renesas.com>
1281 * config/tc-m32r.c (md_convert_frag): Changed for @PLT.
1282 (m32r_cgen_record_fixup_exp): Changed for @GOTOFF, @GOT.
1283 (m32r_fix_adjustable): Changed for @GOTOFF, @GOT, @PLT.
1284 (tc_gen_reloc): Likewise.
1285 (m32r_end_of_match): Add for @GOTOFF, @GOT, @PLT.
1286 (m32r_parse_name): Likewise.
1287 (m32r_cgen_parse_fix_exp): Likewise.
1288 * config/tc-m32r.h (md_parse_name): Define for @GOTOFF, @GOT, @PLT.
1289 (O_PIC_reloc): Likewise.
1290 (TC_CGEN_PARSE_FIX_EXP): Likewise..
1291 * cgen.c (gas_cgen_parse_operand): Add TC_CGEN_PARSE_FIX_EXP
1292 for @GOTOFF, @GOT, @PLT.
1294 2004-06-21 Jan Beulich <jbeulich@novell.com>
1296 * gas/symbols.c: While discarding ordinary local absolute symbols
1297 when --strip-local-absolute is in effect, retain file symbols.
1299 2004-06-20 Andreas Schwab <schwab@suse.de>
1301 * config/tc-m68k.c (mri_chip): Replace current_chip, not augment.
1302 (md_parse_option): Likewise.
1304 2004-06-17 Jan Beulich <jbeulich@novell.com>
1306 * config/tc-i386.c: Deal with LEX_QM the same way as with LEX_AT.
1307 * config/te-netware.h: New file.
1308 * config/te-ppcnw.h: Delete: Obsolete.
1309 * configure.in: Eliminate ill NetWare targets. Make generic
1310 NetWare target use proper emulation.
1311 * Makefile.am: Eliminate reference to obsolete te-ppcnw.h, add
1312 reference to new te-netware.h.
1313 * configure: Regenerate.
1314 * Makefile.in: Regenerate.
1316 2004-06-15 Martin Schwidefsky <schwidefsky@de.ibm.com>
1318 * config/tc-s390.c (s390_insn): Avoid incorrect signed/unsigned
1319 comparison in .insn pseudo operation.
1321 2004-06-15 Alan Modra <amodra@bigpond.net.au>
1323 * config/obj-coff.c (coff_adjust_section_syms): Use
1324 bfd_get_section_size instead of bfd_get_section_size_before_reloc.
1325 (coff_frob_section): Likewise.
1326 * config/tc-mips.c (md_apply_fix3): Likewise.
1327 * config/obj-elf.c (elf_frob_file): Use bfd_set_section_size.
1328 (elf_frob_file_after_relocs): Likewise.
1330 2004-06-10 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
1332 * config/tc-hppa.c (log2): Rename to exact_log2.
1333 (pa_next_subseg): Delete unused function.
1334 (create_new_space): Mark unused arguments with ATTRIBUTE_UNUSED.
1335 (create_new_subspace): Likewise.
1338 * config/tc-hppa.c (hppa_fix_adjustable): Allow reduction of fake
1339 labels. Fix warning.
1341 2004-05-28 DJ Delorie <dj@redhat.com>
1343 * config/tc-mn10300.h (tc_fix_adjustable): Define.
1344 * config/tc-mn10300.c (mn10300_fix_adjustable): Don't adjust debug
1345 or non-merged symbols.
1347 2004-05-28 H.J. Lu <hongjiu.lu@intel.com>
1349 * config/tc-ia64.c (remove_marked_resource): Save, clear and
1350 restore the old slot when inserting srlz.i/srlz.d.
1352 2004-05-28 Andrew Stubbs <andrew.stubbs@superh.com>
1354 * Makefile.am: Regenerate dependecies.
1355 * Makefile.in: Regenerate.
1356 * config/tc-sh.c (valid_arch): Make unsigned.
1357 (preset_target_arch): Likewise.
1358 (md_begin): Use new architecture flags system.
1359 (get_specific): Likewise.
1360 (assemble_ppi): Likewise.
1361 (md_assemble): Likewise. Also fix error check for bad opcodes.
1362 (md_parse_option): Likewise. Also generate -isa values according
1363 to the table in bfd/cpu-sh.c instead of just constants. Also
1364 allow <arch>-up ISA variants.
1365 (sh_elf_final_processing): Replace if-else chain with a call to
1366 sh_find_elf_flags().
1367 * testsuite/gas/sh/arch: New directory.
1368 * testsuite/gas/sh/arch/arch.exp: New test script.
1369 * testsuite/gas/sh/arch/arch_expected.txt: New file.
1370 * testsuite/gas/sh/arch/sh.s: New file.
1371 * testsuite/gas/sh/arch/sh2.s: New file.
1372 * testsuite/gas/sh/arch/sh-dsp.s: New file.
1373 * testsuite/gas/sh/arch/sh2e.s: New file.
1374 * testsuite/gas/sh/arch/sh3-nommu.s: New file.
1375 * testsuite/gas/sh/arch/sh3.s: New file.
1376 * testsuite/gas/sh/arch/sh3-dsp.s: New file.
1377 * testsuite/gas/sh/arch/sh3e.s: New file.
1378 * testsuite/gas/sh/arch/sh4-nommu-nofpu.s: New file.
1379 * testsuite/gas/sh/arch/sh4-nofpu.s: New file.
1380 * testsuite/gas/sh/arch/sh4.s: New file.
1381 * testsuite/gas/sh/arch/sh4a-nofpu.s: New file.
1382 * testsuite/gas/sh/arch/sh4al-dsp.s: New file.
1383 * testsuite/gas/sh/arch/sh4a.s: New file.
1385 2004-05-28 Peter Barada <peter@the-baradas.com>
1387 * config/gc-m68k.c(m68k_ip): Convert mode 5 addressing
1388 with zero offset into mode 2 addressing to save a word.
1390 2004-05-27 H.J. Lu <hongjiu.lu@intel.com>
1392 * config/tc-ia64.c (ar_is_in_integer_unit): Removed.
1393 (ar_is_only_in_integer_unit): New.
1394 (ar_is_only_in_memory_unit): New.
1395 (generate_unwind_image): Silence gcc on 32bit host.
1396 (md_assemble): Use ar_is_only_in_integer_unit instead of
1397 ar_is_in_integer_unit. Check AR access.
1399 2004-05-27 Peter Barada <peter@the-baradas.com>
1401 * config/tc-m68k.c (md_begin): Sort the opcode table into
1403 (m68k_compare_opcode): New function to do the sorting.
1405 2004-05-24 Peter Barada <peter@the-baradas.com>
1407 * config/m68k-parse.y(operand): Allow for MAC/EMAC mask
1408 addressing on MIT style operands.
1409 * config/m68k-parse.y(yylex): Allow '-&' for predecrement
1411 * config/tc-m68k.c(install_operand): Comment 'G' and 'H' type
1414 2004-05-23 Alan Modra <amodra@bigpond.net.au>
1416 * expr.c (operand, operator): Don't reject '++' and '--'.
1418 2004-05-20 Richard Sandiford <rsandifo@redhat.com>
1420 * config/tc-mips.c (append_insn): Use ISA-encoded addresses in MIPS16
1423 2004-05-17 Adam Nemet <anemet@lnxw.com>
1425 * configure.in: Add ppc-*-lynxos*. Update i386-*-lynxos* to ELF.
1426 * configure: Regenerate.
1428 2004-05-13 Paul Brook <paul@codesourcery.com>
1430 * dw2gencfi.c (output_cie): Handle dwarf3 format CIE entries.
1432 2004-05-13 Joel Sherrill <joel@oarcorp.com>
1434 * configure.in (or32-*-rtems*): Switch to elf.
1435 * configure: Regenerate.
1437 2004-05-13 Nick Clifton <nickc@redhat.com>
1439 * po/fr.po: Updated French translation.
1441 2004-05-11 Nick Clifton <nickc@redhat.com>
1443 * doc/as.texinfo (Section): Document G and T flags to .section
1444 directive. Document the extra arguments that the G flag
1445 requires. Document the #tls flag.
1447 2004-05-11 H.J. Lu <hongjiu.lu@intel.com>
1449 * subsegs.c (section_symbol): Create a new section symbol if
1450 the existing one doesn't match.
1451 * symbols.c (symbol_set_bfdsym): Don't reset BFD section symbol.
1453 2004-05-07 Richard Sandiford <rsandifo@redhat.com>
1455 * config/tc-mips.c (append_insn, mips_emit_delays): Extend -mfix-vr4120
1456 to cope with VR4181A errata MD(1) and MD(4).
1458 2004-05-07 Brian Ford <ford@vss.fsi.com>
1460 * NEWS: Mention .secrel32 for pe[i]-i386.
1462 2004-05-07 Alexandre Oliva <aoliva@redhat.com>
1464 * config/tc-frv.h (MAX_MEM_FOR_RS_ALIGN_CODE): New.
1465 (HANDLE_ALIGN): New.
1467 2004-05-06 Daniel Jacobowitz <dan@debian.org>
1469 * Makefile.am (DIST_SUBDIRS): Define.
1470 * aclocal.m4: Regenerate with automake 1.8.4.
1471 * Makefile.in: Likewise.
1472 * doc/Makefile.in: Likewise.
1474 2004-05-06 David Mosberger-Tang <davidm@hpl.hp.com>
1476 * config/tc-ia64.c (dot_serialize): Declare.
1477 (dot_serialize): New function.
1478 (md_pseudo_table): Add ".serialize.data" and
1479 ".serialize.instruction" directives.
1481 2004-05-06 Nick Clifton <nickc@redhat.com>
1483 * messages (as_internal_value_out_of_range): Print a message about
1484 a value being out of range. Be consistent about whether the
1485 values are printed in decimal or hexadecimal.
1486 (as_warn_value_out_of_range): Generate a warning message about an
1488 (as_bad_value_out_of_range): Generate an error message about an
1490 * as.h: Prototype the new functions.
1491 * config/tc-alpha.c (insert_operand): Use new function.
1492 * config/tc-arc.c (arc_insert_operand): Likewise.
1493 * config/tc-mn10200.c (mn10200_insert_operand): Likewise.
1494 * config/tc-mn10300.c (mn10300_insert_operand): Likewise.
1495 * config/tc-ppc.c (ppc_insert_operand): Likewise.
1496 * config/tc-s390.c (s390_insert_operand): Likewise.
1497 * config/tc-v850.c (v850_insert_operand): Likewise.
1499 2004-05-05 Alexandre Oliva <aoliva@redhat.com>
1501 * configure.in: Set em=linux for frv-*-*linux*.
1502 * configure: Rebuilt.
1503 * config/tc-frv.h (TARGET_FORMAT): Use elf32-frvfdpic if...
1504 (frv_md_fdpic_enabled): New.
1505 * config/tc-frv.c (frv_md_fdpic_enabled): New.
1506 (DEFAULT_FDPIC): New.
1507 (frv_flags): Use DEFAULT_FDPIC.
1508 (frv_pic_flag): Likewise.
1509 (OPTION_NOPIC): New.
1510 (md_longopts): Add -mnopic.
1511 (md_parse_option): Handle it.
1512 (md_show_usage): Add -mfdpic and -mnopic.
1514 2004-05-05 Peter Barada <peter@the-baradas.com>
1516 * config/tc-m68k.c: Add find_cf_chip to print list of valid
1517 chips for invalid coldfire instructions, rename selectors
1518 for ColdFire sub-variants, add 521x,5249,547x,548x and aliases,
1519 add current_chip to track which chip is referred to(including save/restore),
1520 use current_chip to select control registers, not current_arch.
1521 (md_show_usage): Add new chips.
1522 * doc/c-m68k.texi: Document new command line switches.
1524 2004-05-05 Jakub Jelinek <jakub@redhat.com>
1526 * tc-s390.h (md_do_align, HANDLE_ALIGN): Remove.
1527 (NOP_OPCODE): Define.
1528 (s390_align_code): Remove prototype.
1529 * tc-s390.c (s390_align_code): Remove.
1531 2004-05-04 H.J. Lu <hongjiu.lu@intel.com>
1533 * config/tc-ia64.c (make_unw_section_name): Removed.
1534 (start_unwind_section): New function.
1535 (generate_unwind_image): Take const segT instead of const
1537 (dot_handlerdata): Adjusted.
1538 (dot_endp): Likewise.
1540 2004-05-02 H.J. Lu <hongjiu.lu@intel.com>
1542 * config/obj-elf.c (obj_elf_change_section): Allow the
1543 ".note.GNU-stack" section has SHF_EXECINSTR.
1545 2004-05-02 H.J. Lu <hongjiu.lu@intel.com>
1547 * config/obj-elf.c (get_section): Return bfd_boolean.
1548 (obj_elf_change_section): Call bfd_get_section_by_name_if
1549 instead of bfd_map_over_sections.
1551 2004-04-30 H.J. Lu <hongjiu.lu@intel.com>
1553 * config/obj-elf.c (get_section): New function.
1554 (obj_elf_change_section): Support multiple sections with same
1557 2004-04-30 Nick Clifton <nickc@redhat.com>
1559 * config/tc-arm.c (create_register_alias): Fix typo checking for
1560 case sensitive register aliases.
1561 (co_proc_number): Use error message string in all_reg_maps[]
1563 (cp_reg_required_here): Likewise.
1564 (fp_reg_required_here): Likewise.
1566 2004-04-29 Brian Ford <ford@vss.fsi.com>
1568 * dwarf2dbg.c (dwarf2_finish): Add SEC_DEBUGGING to section flags.
1570 2004-04-28 Chris Demetriou <cgd@broadcom.com>
1572 * config/tc-mips.c (HAVE_32BIT_ADDRESSES, append_insn, macro_build)
1573 (load_address, macro, mips_ip, md_parse_option)
1574 (mips_force_relocation, mips_validate_fix, md_apply_fix3)
1575 (s_change_sec, pic_need_relax, tc_gen_reloc): Remove all
1576 embedded-PIC handling, and update comments.
1577 (SWITCH_TABLE): Remove.
1578 * config/tc-mips.h (DIFF_EXPR_OK): Delete.
1579 (enum mips_pic_level): Remove EMBEDDED_PIC.
1580 (EXTERN_FORCE_RELOC): Remove embedded-PIC handling.
1581 (TC_FORCE_RELOCATION): Update comment.
1582 * ecoff.c (ecoff_build_lineno): Add comment about some code that
1583 might be safe to remove now that MIPS embedded-PIC is gone.
1585 2004-04-28 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
1587 * config/obj-som.c (obj_som_init_stab_section): Add new arguments in
1588 call to obj_set_subsection_attributes.
1589 (obj_som_init_stab_section): Likewise.
1590 * config/tc-hppa.c (default_subspace_dict): Add comdat field.
1591 (pa_def_subspaces): Provide comdat default.
1592 (pa_subspace): Handle new "comdat" parameter. Set SEC_LINK_ONCE and
1593 not SEC_IS_COMMON if section is comdat, common or dup_common. Update
1594 calls to create_new_subspace and update_subspace to pass comdat flag.
1595 (create_new_subspace, update_subspace): Add new comdat argument. Use
1596 it in calls to obj_set_subsection_attributes.
1597 * doc/c-hppa.texi (.subspa, .nsubspa): Document new comdat parameter
1598 and use of comdat, common and dup_comm parameters.
1600 2004-04-26 H.J. Lu <hongjiu.lu@intel.com>
1602 * config/obj-elf.c (obj_elf_change_section): Check if the old
1603 group name is NULL before comparison.
1605 2004-04-23 Chris Demetriou <cgd@broadcom.com>
1607 * config/tc-mips.h (mips_dwarf2_addr_size): Prototype.
1609 2004-04-23 Thiemo Seufer <seufer@csv.ica.uni-stuttgart.de>
1611 * config/tc-mips.c (s_mipsset): Set default CPU type for .set mipsN.
1613 2004-04-23 Chris Demetriou <cgd@broadcom.com>
1615 * config/tc-mips.c (md_longopts): Remove -membedded-pic option.
1616 (OPTION_MEMBEDDED_PIC): Remove.
1617 (OPTION_TRAP, OPTION_BREAK, OPTION_EB, OPTION_EL)
1618 (OPTION_FP32, OPTION_GP32, OPTION_CONSTRUCT_FLOATS)
1619 (OPTION_NO_CONSTRUCT_FLOATS, OPTIONS_FP64, OPTION_GP64)
1620 (OPTION_RELAX_BRANCH, OPTION_NO_RELAX_BRANCH)
1621 (OPTION_ELF_BASE): Renumber.
1622 (md_parse_option): Remove OPTION_MEMBEDDED_PIC handling.
1623 (md_show_usage): Remove mention of -membedded-pic.
1624 * doc/as.texinfo: Remove mention of -membedded-pic.
1626 2004-04-23 Thiemo Seufer <seufer@csv.ica.uni-stuttgart.de>
1628 * config/tc-mips.h (USE_GLOBAL_POINTER_OPT): Remove.
1629 * config/tc-mips.c (RDATA_SECTION_NAME, mips_target_format): Remove
1631 (md_begin, mips_ip, md_parse_option, s_change_sec, s_option,
1632 s_abicalls, nopic_need_relax, tc_gen_reloc): Remove uses of
1633 USE_GLOBAL_POINTER_OPT.
1635 2004-04-22 Thiemo Seufer <seufer@csv.ica.uni-stuttgart.de>
1637 * config/tc-mips.c (macro): One more use of load_delay_nop.
1639 2004-04-22 Atsushi Nemoto <anemo@mba.ocn.ne.jp>
1641 * config/tc-mips.c (load_delay_nop): New function.
1642 (load_address, macro): Use load_delay_nop() to build a nop
1643 which can be omitted with gpr_interlocks.
1645 2004-04-22 Thiemo Seufer <seufer@csv.ica.uni-stuttgart.de>
1647 * config/tc-mips.c (hilo_interlocks, gpr_interlocks,
1648 cop_interlocks): Remove superfluous CPU entries.
1650 2004-04-22 Paul Brook <paul@codesourcery.com>
1652 * config/tc-arm.c (mav_parse_offset): Value must be multiple of 4.
1654 2004-04-22 Peter Barada <peter@the-baradas.com>
1656 * NEWS: Added support for EMAC instructions and MAC/EMAC
1658 * config/m68k-parse.h: Add ACC[123], ACCEXT{01,23}, MAC/EMAC
1659 scale factor tokens, trailing_ampersand to mark mask addressing
1660 for MAC/EMAC instructions.
1661 * config/m68k-parse.y: Add options_ampersand clause, '<<',
1663 (yylex): Handle '>', '<', and '&' following '+'.
1664 * config/tc-m68k.c: Set mcfmac/mcfemac on appropriate ColdFire
1665 architectures in archs[].
1666 (m68k-ip): Add '4', 'e', 'g', 'i', cases to handle mask addressing
1667 for MAC/EMAC instructions, ACC[0123], ACCEXT{01,23}, and '<<'/'>>'
1669 (m68k_ip): Handle trailing '&' on MAC/EMAC insns.
1670 (install_operand): Fix 'n' case, Add 'F', 'f', 'G', 'H', 'I', ']'
1672 Add EMAC operands to init_table[].
1674 2004-04-22 Bruno De Bus <bdebus@elis.ugent.be>
1676 * config/tc-arm.h (enum mstate): Move here, add MAP_UNDEFINED
1678 (TC_SEGMENT_INFO_TYPE): Define to enum mstate.
1679 * config/tc-arm.c (enum mstate): Delete from here.
1680 (mapping_state): Remove the static mapstate variable and instead
1681 store the state in the segment. This allows a per-section mapping
1682 state. Handle and ignore MAP_UNDEFINED states.
1683 (arm_elf_change_section): Get the current mapping state from the
1685 (s_ltorg): Set the mapping state to MAP_DATA.
1686 (arm_cleanup): Use arm_elf_change_section to get the mapping state
1687 for each pool as it is emitted.
1689 2004-04-22 Nick Clifton <nickc@redhat.com>
1691 * config/tc-arm.h: Formatting tidy ups.
1693 2004-04-20 Chris Demetriou <cgd@broadcom.com>
1695 * NEWS: Note that MIPS -membedded-pic option is deprecated.
1697 2004-04-20 DJ Delorie <dj@redhat.com>
1699 * config/tc-i386.h [TE_PE] (TC_CONS_FIX_NEW): Define.
1700 * config/tc-i386.c (md_pseudo_table) [TE_PE]: Add "secrel32".
1701 [TE_PE] (O_secrel): Define.
1702 [TE_PE] (x86_pe_cons_fix_new): New.
1703 [TE_PE] (pe_directive_secrel): Likewise.
1704 (tc_gen_reloc) [TE_PE]: Support BFD_RELOC_32_SECREL.
1706 2004-04-19 Eric Christopher <echristo@redhat.com>
1708 * config/tc-mips.c (mips_dwarf2_addr_size): Revert part
1709 of previous patch for fix in gcc.
1711 2004-04-19 Jakub Jelinek <jakub@redhat.com>
1713 * config/tc-xtensa.c (xg_assembler_literal): Fix a typo.
1715 2004-04-19 Nathan Sidwell <nathan@codesourcery.com>
1717 * read.c (do_align): Call md_flush_pending_output, if defined.
1719 2004-04-16 Alan Modra <amodra@bigpond.net.au>
1721 * expr.c (operand): Correct checks for ++ and --.
1723 2004-04-14 H.J. Lu <hongjiu.lu@intel.com>
1725 * config/tc-generic.c: Add some comments.
1727 2004-04-14 Richard Sandiford <rsandifo@redhat.com>
1729 * doc/c-mips.texi (-m{no-,}fix-vr4120): Renamed from
1730 -{no-}mfix-vr4122-bugs.
1731 * config/tc-mips.c (mips_fix_vr4120): Renamed from mips_fix_4122_bugs.
1732 (append_insn, mips_emit_delays): Update accordingly.
1733 (OPTION_FIX_VR4120, OPTION_NO_FIX_VR4120): Renamed from *VR4122.
1734 (md_longopts): Change -{no-,}mfix-vr4122-bugs to -m{no-,}fix-vr4120.
1735 (md_parse_option): Update after above changes.
1736 (md_show_usage): Add -mfix-vr4120.
1738 2004-04-13 Bob Wilson <bob.wilson@acm.org>
1740 * doc/as.texinfo (Sub-Sections): Conditionalize COFF-specific use
1741 of .section directive; add a reference to the ELF .subsection
1744 2004-04-13 Kazuhiro Inaoka <inaoka.kazuhiro@renesas.com>
1746 * config/tc-m32r.c (md_assemble): Fixed infinite loop bug
1749 2004-04-11 Thiemo Seufer <seufer@csv.ica.uni-stuttgart.de>
1751 * Makefile.am: Remove mips from aout targets.
1752 * Makefile.in: Regenerate.
1753 * configure.in: Remove mips-dec-bsd* target.
1754 * configure: Regenerate.
1756 2004-04-07 Alan Modra <amodra@bigpond.net.au>
1759 * config/tc-ppc.c (ppc_elf_suffix): Add valid32 and valid64 fields
1760 to struct map_bfd. Adjust MAP macro, and define MAP32, MAP64.
1761 Update "mapping". Restrict some @ modifiers to 32 bit.
1763 2004-04-01 Asgari Jinia <asgarij@kpitcummins.com>
1764 Dhananjay Deshpande <dhananjayd@kpitcummins.com>
1766 * config/tc-sh.c (dont_adjust_reloc_32): New variable.
1767 (sh_fix_adjustable): Avoid adjusting BFD_RELOC_32 when
1768 dont_adjust_reloc_32 is set.
1769 (md_longopts): Add option -renesas.
1770 (md_parse_option, md_show_usage): Likewise.
1771 * doc/c-sh.texi: Likewise.
1773 2004-04-01 Dave Korn <dk@artimi.com>
1775 * config/tc-dlx.c (md_assemble): set fx_no_overflow flag for
1776 hi16 and lo16 fixS structs.
1777 (md_assemble): generate bit_fixS for RELOC_DLX_LO16 in
1778 exactly the same way as for RELOC_DLX_REL16.
1779 (machine_ip): properly respect LO flag in the_insn and
1780 output RELOC_DLX_LO16 rather than RELOC_DLX_16.
1781 (md_apply_fix3): apply RELOC_DLX_LO16.
1783 2004-03-30 Stan Shebs <shebs@apple.com>
1785 Remove long-obsolete MPW support.
1786 * mpw-config.in, mpw-make.sed, mac-as.r: Remove files.
1787 * configure.in: Remove mention of ppc-*-mpw* config.
1788 * configure.in: Likewise.
1790 2004-03-30 Nick Clifton <nickc@redhat.com>
1792 * config/tc-arm.c (meabi_flags): Make its use conditional upon
1793 OBJ_ELF being defined.
1795 2004-03-27 Alan Modra <amodra@bigpond.net.au>
1797 * config/obj-aout.c (obj_aout_type): Remove #ifdef BFD_ASSEMBLER code.
1799 2004-03-23 Paul Brook <paul@codesourcery.com>
1801 * config/tc-arm.c (meabi_flags): New variable.
1802 (arm_parse_eabi): New function.
1803 (md_begin): Set flags for EABI v3.
1805 (arm_long_opts): Add meabi.
1806 * doc/as.texinf <ARM>: Document -meabi.
1807 * doc/c-arm.texi: Ditto.
1809 2004-03-22 Bob Wilson <bob.wilson@acm.org>
1811 * config/tc-xtensa.c (xtensa_post_relax_hook): Create literal
1812 tables even when use_literal_section flag is not set.
1814 2004-03-22 Alan Modra <amodra@bigpond.net.au>
1816 * config/tc-sh.c: Remove trailing whitespace.
1818 2004-03-22 Hans-Peter Nilsson <hp@axis.com>
1820 * doc/c-cris.texi (CRIS-Opts): Document --no-mul-bug-abort,
1821 --mul-bug-abort and the default behavior.
1822 * config/tc-cris.c (cris_insn_kind): New member CRIS_INSN_MUL.
1823 (err_for_dangerous_mul_placement): New variable.
1824 (STATE_MUL, OPTION_MULBUG_ABORT_ON, OPTION_MULBUG_ABORT_OFF): New
1826 (md_cris_relax_table): Have placeholder for STATE_MUL.
1827 (md_longopts): New options --mul-bug-abort and --no-mul-bug-abort.
1828 (cris_relax_frag) <case ENCODE_RELAX (STATE_MUL, STATE_BYTE)>: New
1830 (md_estimate_size_before_relax) <case ENCODE_RELAX (STATE_MUL,
1831 STATE_BYTE)>: Ditto.
1832 (md_convert_frag) <ENCODE_RELAX (STATE_MUL, STATE_BYTE)>: Check
1833 alignment and position of this frag, emit error message if
1835 (md_assemble): For a multiply insn and when checking it,
1836 transform the current frag into a special frag for that purpose.
1837 (md_parse_option) <case OPTION_MULBUG_ABORT_OFF, case
1838 OPTION_MULBUG_ABORT_ON>: Handle new options.
1840 2004-03-19 Bob Wilson <bob.wilson@acm.org>
1842 * config/tc-xtensa.c (mark_literal_frags): New function.
1843 (xtensa_move_literals): Call mark_literal_frags for all literal
1844 segments, including init and fini literal segments.
1845 (xtensa_post_relax_hook): Swap use of xt_insn_sec and xt_literal_sec.
1847 2004-03-19 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
1849 * tc-hppa.c (cons_fix_new_hppa): Check for PC relative base type.
1850 (pa_comm): Set BSF_OBJECT in symbol flags.
1852 2004-03-19 Alan Modra <amodra@bigpond.net.au>
1854 * Makefile.am: Run "make dep-am".
1855 * Makefile.in: Regenerate.
1856 * doc/Makefile.in: Regenerate.
1857 * config.in: Regenerate.
1858 * po/gas.pot: Regenerate.
1860 2004-03-18 Nathan Sidwell <nathan@codesourcery.com>
1862 * read.c (read_a_source_file): Use demand_empty_rest_of_line.
1863 (demand_empty_rest_of_line): Issue an error here.
1864 (ignore_rest_of_line): Silently skip to end.
1865 (demand_copy_string): Issue an error, not warning.
1867 * config/obj-elf.c (obj_elf_section_name): Likewise.
1868 (obj_elf_section): Likewise.
1869 * config/tc-arc.c (arc_extoper): Remove bogus NULL checks.
1870 (arc_extinst): Likewise.
1871 * config/tc-ia64.c (dot_saveb): Use demand_empty_rest_of_line.
1872 (dot_spill): Likewise.
1873 (dot_unwabi): Likewise.
1874 (dot_prologue): Likewise.
1876 * expr.c (operand): Reject ++ and --.
1877 (operator): Likewise.
1879 2004-03-17 Kaz Kojima <kkojima@rr.iij4u.or.jp>
1881 * config/tc-sh.c: Include dw2gencfi.h.
1882 (sh_cfi_frame_initial_instructions): New function.
1883 (sh_regname_to_dw2regnum): Likewise.
1884 * config/tc-sh.h (DWARF2_LINE_MIN_INSN_LENGTH): Move to the end of
1886 (TARGET_USE_CFIPOP): Define.
1887 (tc_cfi_frame_initial_instructions): Likewise.
1888 (tc_regname_to_dw2regnum): Likewise.
1889 (DWARF2_DEFAULT_RETURN_COLUMN, DWARF2_CIE_DATA_ALIGNMENT): Likewise.
1890 * Makefile.am: Update dependencies.
1891 * Makefile.in: Regenerate.
1893 2004-03-17 Ralf Corsepius <corsepiu@faw.uni-ulm.de>
1895 * configure.in: Switch sh-*-rtems* to ELF. Add sh-*-rtemscoff*.
1896 * configure: Regenerate.
1898 2004-03-12 Bob Wilson <bob.wilson@acm.org>
1900 * read.c (s_leb128): Call md_flush_pending_output.
1902 2004-03-12 Michal Ludvig <mludvig@suse.cz>
1904 * config/tc-i386.c (output_insn): Handle PadLock instructions.
1905 * config/tc-i386.h (CpuPadLock): New define.
1906 (CpuUnknownFlags): Added CpuPadLock.
1908 2004-03-07 Andreas Schwab <schwab@suse.de>
1910 * doc/c-hppa.texi (HPPA Directives): Fix typo.
1912 2004-03-07 Richard Henderson <rth@redhat.com>
1914 * dw2gencfi.c (output_cie): Align length to 4 byte boundary.
1915 (cfi_finish): Likewise for fde.
1917 2004-03-05 H.J. Lu <hongjiu.lu@intel.com>
1919 * config/tc-ia64.c (md_assemble): Properly handle NULL
1921 (ia64_handle_align): Don't abort if failed to add a stop bit.
1923 2004-03-04 H.J. Lu <hongjiu.lu@intel.com>
1925 * Makefile.in: Regenerated.
1926 * aclocal.m4: Likewise.
1927 * configure: Likewise.
1928 * doc/Makefile.in: Likewise.
1930 2004-03-03 H.J. Lu <hongjiu.lu@intel.com>
1932 * config/tc-ia64.c (dot_align): New.
1933 (ia64_do_align): Make it static.
1934 (md_pseudo_table): Use "dot_align" for "align".
1935 (ia64_md_do_align): Don't set align_frag here.
1936 (ia64_handle_align): Add a stop bit to the previous bundle if
1939 * config/tc-ia64.h (ia64_do_align): Removed.
1941 2003-03-03 Andrew Stubbs <andrew.stubbs@superh.com>
1943 * config/tc-sh.c (md_parse_option): Add -isa=sh4-nofpu and
1944 -isa=sh4-nommu-nofpu options. Adjust help messages accordingly.
1945 (sh_elf_final_processing): Output BFD type sh4_nofpu if that is
1946 the most general type or the user specifically requested it.
1947 (md_assemble): Add a new error message for when an instruction
1948 is understood, but is not allowed due to an -isa option.
1950 2004-03-02 H.J. Lu <hongjiu.lu@intel.com>
1952 * config/tc-ia64.c (align_frag): New.
1953 (md_assemble): Set the tc_frag_data field in align_frag for
1954 IA64_OPCODE_FIRST instructions.
1955 (ia64_md_do_align): Set align_frag.
1956 (ia64_handle_align): Add a stop bit if needed.
1958 * config/tc-ia64.h (TC_FRAG_TYPE): New.
1959 (TC_FRAG_INIT): New.
1961 2004-03-01 Richard Sandiford <rsandifo@redhat.com>
1963 * config/tc-frv.c (fr400_audio): New variable.
1964 (md_parse_option, md_show_usage): Add -mcpu=fr405 and -mcpu=fr450.
1965 (md_parse_option): Set fr400_audio for -mcpu=fr400 and -mcpu=fr405.
1966 (target_implements_insn_p): New function.
1967 (md_assemble): Report an error if the processor doesn't implement
1970 2004-02-27 Kazuhiro Inaoka <inaoka.kazuhiro@renesas.com>
1972 * config/tc-m32r.c (md_longopts): Added -no-bitinst option.
1973 (md_parse_option): Ditto.
1974 (OPTION_NO_SPECIAL_M32R): Added.
1975 (md_show_usage): Document it.
1976 (enable_speial_m32r): Changed a default value from 0 to 1.
1977 * doc/c-m32r.texi: Document the -no-bitinst option.
1979 2004-02-27 Nick Clifton <nickc@redhat.com>
1981 * config/tc-sh.c (get_operand): Revert previous delta.
1982 (tc_gen_reloc): Check for an unknown reloc type before processing
1985 2004-02-27 Hannes Reinecke <hare@suse.de>
1987 * config/tc-s390.c (s390_insn): Correct range check for opcode in
1988 .insn pseudo operation.
1990 2004-02-27 Anil Paranjpe <anilp1@kpitcummins.com>
1992 * config/tc-sh.c (get_operand): In case of #Imm, check has been
1993 added for wrong syntax.
1995 2004-02-26 Eric Christopher <echristo@redhat.com>
1997 * config/tc-mips.c (mips_dwarf2_addr_size): New.
1998 * config/tc-mips.h (DWARF2_ADDR_SIZE): Use.
2000 2004-02-26 Andrew Stubbs <andrew.stubbs@superh.com>
2002 * config/tc-sh.c (build_Mytes): Add REG_N_D and REG_N_B01
2003 nibble types to assembler.
2005 2004-02-25 Fred Fish <fnf@redhat.com>
2007 * config/tc-iq2000.c: Add missing \n\ in multiline string literal.
2009 2004-02-20 James E Wilson <wilson@specifixinc.com>
2011 * config/tc-ia64.c (slot_index): New arg before_relax. Use instead of
2013 (fixup_unw_records): New arg before_relax. Pass to slot_index.
2014 (ia64_estimate_size_before_relax): New.
2015 (ia64_convert_frag): Pass 0 to fixup_unw_records. Add comment.
2016 (generate_unwind_image): Pass 1 to fixup_unw_records.
2017 * config/tc-ia64.h (ia64_estimate_size_before_relax): Declare.
2018 (md_estimate_size_before_relax): Call ia64_estimate_size_before_relax.
2020 2004-02-19 Jakub Jelinek <jakub@redhat.com>
2022 * stabs.c (generate_asm_file): Avoid warning about use of
2023 uninitialized variable.
2025 2004-02-18 David Mosberger <davidm@hpl.hp.com>
2027 * config/tc-ia64.c (ia64_flush_insns): In addition to prologue,
2028 body, and endp, allow unwind records which do not have a "t"
2029 (time/instruction) field.
2031 2004-02-17 Petko Manolov <petkan@nucleusys.com>
2033 * config/tc-arm.c (do_mav_dspsc_1): Correct offset of CRn.
2034 (do_mav_dspsc_2): Likewise.
2035 Fix accumulator registers move opcodes.
2037 2004-02-13 Hannes Reinecke <hare@suse.de>
2038 Jakub Jelinek <jakub@redhat.com>
2040 * dwarf2dbg.c (get_filenum): Do not read beyond allocated memory.
2042 2004-02-10 Steve Ellcey <sje@cup.hp.com>
2044 * config/tc-ia64.h (ia64_frob_symbol): New declaration.
2045 (tc_frob_symbol): New macro definition.
2046 * config/tc-ia64.c (ia64_frob_symbol): New routine.
2048 2004-02-09 Daniel Jacobowitz <drow@mvista.com>
2050 * config/tc-arm.c (md_begin): Mark .note.gnu.arm.ident as
2053 2004-02-09 Nathan Sidwell <nathan@codesourcery.com>
2055 * read.h (IGNORE_OPCODE_CASE): Do not define. Replace with ...
2056 (TC_CASE_SENSITIVE): ... this.
2057 * read.c: Replace IGNORE_OPCODE_CASE with TC_CASE_SENSITIVE.
2058 * doc/internals.texi (TC_CASE_SENSITIVE): Document.
2060 2004-02-06 James E Wilson <wilson@specifixinc.com>
2062 * config/tc-ia64.c (dot_endp): Delete call to output_endp.
2063 (generate_unwind_image): Re-add it here.
2065 2004-02-06 Nathan Sidwell <nathan@codesourcery.com>
2067 * dwarf2dbg.c (DWARF2_ADDR_SIZE): Remove trailing ';'
2068 * read.h (SKIP_WHITESPACE): Turn into an expression.
2069 * read.c (read_a_source_file): A pseudo is removed by having a
2072 2004-02-05 James E Wilson <wilson@specifixinc.com>
2074 * config/tc-ia64.c (output_endp): New.
2075 (count_bits): Delete.
2076 (ia64_flush_insns, process_one_record, optimize_unw_records): Handle
2077 endp unwind records.
2078 (fixup_unw_records): Handle endp unwind records. Delete code for
2079 shortening prologue regions not followed by a body record.
2080 (dot_endp): Call add_unwind_entry to emit endp unwind record.
2081 * config/tc-ia64.h (unw_record_type): Add endp.
2083 2004-02-03 James E Wilson <wilson@specifixinc.com>
2085 * config/tc-ia64.c (ia64_convert_frag): Call md_number_to_chars to
2086 fill padding bytes with zeroes.
2087 (emit_one_bundle): New locals last_ptr, end_ptr. Rewrite code that
2088 sets unwind_record slot_number and slot_frag fields.
2090 2004-02-02 Maciej W. Rozycki <macro@ds2.pg.gda.pl>
2092 * config/tc-mips.c (add_got_offset_hilo): New function.
2093 (macro): Use load_register() and add_got_offset_hilo() to load
2094 constants instead of hardcoding code sequences throughout.
2096 2004-01-28 H.J. Lu <hongjiu.lu@intel.com>
2098 * config/tc-ia64.c (emit_one_bundle): Add proper indentation.
2100 2004-01-26 Bernardo Innocenti <bernie@develer.com>
2102 * config/tc-m68k.h (EXTERN_FORCE_RELOC): Handle m68k-uclinux specially,
2104 * config/tc-m68k.c (RELAXABLE_SYMBOL): Use EXTERN_FORCE_RELOC instead
2105 of hard-coded test for TARGET_OS=elf.
2107 2004-01-24 Chris Demetriou <cgd@broadcom.com>
2109 * config/tc-mips.c (hilo_interlocks): Change definition
2110 so that MIPS32, MIPS64 and later ISAs are included, along with
2111 the already-included machines. Update comments.
2113 2004-01-23 Daniel Jacobowitz <drow@mvista.com>
2115 * config/tc-arm.c (tc_gen_reloc): Improve error message for
2116 undefined local labels.
2118 2004-01-23 Richard Sandiford <rsandifo@redhat.com>
2120 * config/tc-mips.c (load_address, macro): Update comments about
2121 NewABI GP relaxation.
2123 2004-01-23 Richard Sandiford <rsandifo@redhat.com>
2125 * config/tc-mips.c (macro_build): Remove place and counter arguments.
2126 (mips_build_lui, macro_build_ldst_constoffset): Likewise.
2127 (mips16_macro_build, macro_build_jalr): Remove counter argument.
2128 (set_at, load_register, load_address, move_register): Likewise.
2129 (load_got_offset, add_got_offset): Likewise.
2130 Update all calls and tidy accordingly.
2132 2004-01-23 Richard Sandiford <rsandifo@redhat.com>
2134 * config/tc-mips.c (RELAX_ENCODE): Remove WARN argument.
2135 (RELAX_FIRST, RELAX_SECOND): Turn into 8-bit quantities.
2136 (RELAX_USE_SECOND): Bump to 0x10000.
2137 (RELAX_SECOND_LONGER, RELAX_NOMACRO, RELAX_DELAY_SLOT): New flags.
2138 (mips_macro_warning): New variable.
2139 (md_assemble): Wrap macro expansion in macro_start() and macro_end().
2140 (s_cpload, s_cpsetup, s_cprestore, s_cpreturn): Likewise.
2141 (relax_close_frag): Set mips_macro_warning.first_frag. Adjust use
2143 (append_insn): Update mips_macro_warning.sizes.
2144 (macro_start, macro_warning, macro_end): New functions.
2145 (macro_build): Don't emit warnings here.
2146 (macro_build_lui, md_estimate_size_before_relax): ...or here.
2147 (md_convert_frag): Check for cases where one macro alternative
2148 needs a warning and the other doesn't. Emit a warning if the
2149 longer sequence was chosen.
2151 2004-01-23 Richard Sandiford <rsandifo@redhat.com>
2153 * config/tc-mips.h (tc_frag_data_type, TC_FRAG_TYPE): Remove.
2154 * config/tc-mips.c (RELAX_ENCODE): Take three arguments: the size of
2155 the first sequence, the size of the second sequence, and a flag
2156 that says whether we should warn.
2157 (RELAX_OLD, RELAX_NEW, RELAX_RELOC[123]): Delete.
2158 (RELAX_FIRST, RELAX_SECOND): New.
2159 (mips_relax): New variable.
2160 (relax_close_frag, relax_start, relax_switch, relax_end): New fns.
2161 (append_insn): Remove "place" argument. Use mips_relax.sequence
2162 rather than "place" to check whether we're expanding the second
2163 alternative of a relaxable macro. Remove redundant check for
2164 branch relaxation. If generating a normal insn, and there
2165 is not enough room in the current frag, call relax_close_frag()
2166 to close it. Update mips_relax.sizes[]. Emit fixups for the
2167 second version of a relaxable macro. Record the first relaxable
2168 fixup in mips_relax. Remove tc_gen_reloc workaround.
2169 (macro_build): Remove all uses of "place". Use mips_relax.sequence
2170 in the same way as in append_insn.
2171 (mips16_macro_build): Remove "place" argument.
2172 (macro_build_lui): As for macro_build. Don't drop the add_symbol
2173 when generating the second version of a relaxable macro.
2174 (load_got_offset, add_got_offset): New functions.
2175 (load_address, macro): Use new relaxation machinery. Remove
2176 tc_gen_reloc workarounds.
2177 (md_estimate_size_before_relax): Set RELAX_USE_SECOND if the second
2178 version of a relaxable macro is needed. Return -RELAX_SECOND if the
2179 first version is needed.
2180 (tc_gen_reloc): Remove relaxation handling.
2181 (md_convert_frag): Go through the fixups for a relaxable macro and
2182 mark those that belong to the unneeded alternative as done. If the
2183 second alternative is needed, adjust the fixup addresses to account
2184 for the deleted first alternative.
2186 2004-01-23 Richard Sandiford <rsandifo@redhat.com>
2188 * frags.h (frag_room): Declare.
2189 * frags.c (frag_room): New function.
2190 * doc/internals.texi: Document it.
2192 2004-01-22 Thiemo Seufer <seufer@csv.ica.uni-stuttgart.de>
2194 * config/tc-mips.c (append_insn): Don't do r3900 interlock
2195 optimization for -mtune=r3900, as this will break on other CPUs.
2197 2004-01-11 Tom Rix <tcrix@worldnet.att.net>
2199 * config/tc-m68hc11.c (build_indexed_byte): movb and movw cannot
2200 be relaxed, use fixup.
2201 (md_apply_fix3): Use 5 bit reloc from movb and movw fixup.
2203 2004-01-19 Jakub Jelinek <jakub@redhat.com>
2205 * config/tc-sparc.c (sparc_ip): Disallow %f32-%f63 for single
2208 2004-01-14 Maciej W. Rozycki <macro@ds2.pg.gda.pl>
2210 * config/tc-mips.c (append_insn): Properly detect variant frags
2211 that preclude swapping of relaxed branches. Correctly swap
2212 instructions between frags when dealing with relaxed branches.
2214 2004-01-14 Maciej W. Rozycki <macro@ds2.pg.gda.pl>
2216 * acinclude.m4: Quote names of macros to be defined by AC_DEFUN
2218 * aclocal.m4: Regenerate.
2219 * configure: Regenerate.
2221 2004-01-12 Anil Paranjpe <anilp1@KPITCummins.com>
2223 * config/tc-h8300.c (build_bytes): Apply relaxation to bit
2226 2004-01-12 Richard Sandiford <rsandifo@redhat.com>
2228 * config/tc-mips.c (macro_build_jalr): When adding an R_MIPS_JALR
2229 reloc, reserve space for the delay slot as well as the jalr itself.
2231 2004-01-09 Paul Brook <paul@codesourcery.com>
2233 * config/tc-arm.c (do_vfp_reg2_from_sp2): Rename from do_vfp_sp_reg2.
2234 (do_vfp_sp2_from_reg2): New function.
2236 (do_vfp_dp_from_reg2): Check return values properly.
2238 2004-01-08 Ian Lance Taylor <ian@wasabisystems.com>
2240 * config/tc-mips.c (warn_nops): Remove static variable.
2241 (macro): Remove test of warn_nops.
2242 (md_shortops): Remove 'n'.
2243 (md_parse_option): Remove 'n' case.
2244 (md_show_usage): Remove -n.
2245 * doc/as.texinfo (Overview): Remove MIPS -n option.
2246 * doc/c-mips.texi (MIPS Opts): Remove mention -n.
2247 * NEWS: Mention removal of MIPS -n option.
2249 * config/tc-mips.c (ISA_HAS_COPROC_DELAYS): Remove.
2250 (cop_interlocks): Check ISA level.
2251 (cop_mem_interlocks): Define.
2252 (reg_needs_delay): Check cop_interlocks rather than
2253 ISA_HAS_COPROC_DELAYS.
2254 (append_insn): Likewise. Use cop_mem_interlocks rather than
2255 directly checking mips_opts.isa.
2256 (mips_emit_delays): Likewise.
2258 2004-01-07 H.J. Lu <hongjiu.lu@intel.com>
2260 * config/tc-ia64.c (unwind): Move next_slot_number and
2261 next_slot_frag to ...
2262 (unw_rec_list): Here.
2263 (free_list_records): Removed.
2264 (output_unw_records): Likewise.
2265 (generate_unwind_image): Make it void.
2266 (alloc_record): Initialize next_slot_number and next_slot_frag.
2267 (slot_index): Take .org, .space and .align into account.
2268 (fixup_unw_records): Don't set slot_number to 0. Use
2269 list->next_slot_number and list->next_slot_frag instead of
2270 unwind.next_slot_number and unwind.next_slot_frag.
2271 (ia64_convert_frag): New.
2272 (generate_unwind_image): Generate a rs_machine_dependent frag
2274 (emit_one_bundle): Use list->next_slot_number and
2275 list->next_slot_frag instead of unwind.next_slot_number and
2276 unwind.next_slot_frag.
2278 * config/tc-ia64.h (md_convert_frag): Defined as
2280 (md_estimate_size_before_relax): Defined as (f)->fr_var.
2282 2004-01-06 Alexandre Oliva <aoliva@redhat.com>
2284 2003-12-19 Alexandre Oliva <aoliva@redhat.com>
2285 * config/tc-frv.h (md_apply_fix3): Don't define.
2286 * config/tc-frv.c (md_apply_fix3): New. Shift/truncate %hi/%lo
2288 * config/tc-frv.h (TC_FORCE_RELOCATION_SUB_LOCAL): Define.
2289 2003-10-07 Alexandre Oliva <aoliva@redhat.com>
2290 * config/tc-frv.c (line_separator_chars): Add `!'.
2291 2003-09-19 Alexandre Oliva <aoliva@redhat.com>
2292 * config/tc-frv.c (md_assemble): Clear insn upfront.
2293 2003-09-18 Alexandre Oliva <aoliva@redhat.com>
2294 * config/tc-frv.c (OPTION_FDPIC): New macro.
2295 (md_longopts): Add mfdpic.
2296 (md_parse_option): Handle it.
2297 2003-08-04 Alexandre Oliva <aoliva@redhat.com>
2298 * config/tc-frv.c (md_cgen_lookup_reloc) <FRV_OPERAND_D12,
2299 FRV_OPERAND_S12>: Use reloc type encoded in fix-up.
2300 (frv_pic_ptr): Parse funcdesc.
2302 2004-01-05 Maciej W. Rozycki <macro@ds2.pg.gda.pl>
2304 * doc/as.texinfo: Let texi2pod parse asconfig.texi and
2305 gasver.texi. Remove duplicate symbol definitions for texi2pod.
2307 2004-01-05 Maciej W. Rozycki <macro@ds2.pg.gda.pl>
2309 * Makefile.am (Makefile): Move the dependency on
2310 $(BFDDIR)/configure.in to...
2311 (CONFIG_STATUS_DEPENDENCIES): ... here.
2312 (AUTOMAKE_OPTIONS): Require automake 1.8.
2313 * Makefile.in: Regenerate.
2314 * doc/Makefile.am (BASEDIR, BFDDIR): Define.
2315 (CONFIG_STATUS_DEPENDENCIES): Add a dependency on
2316 $(BFDDIR)/configure.in here as well.
2317 * doc/Makefile.in: Regenerate.
2319 2004-01-05 Maciej W. Rozycki <macro@ds2.pg.gda.pl>
2321 * Makefile.am (install, install-info, RECURSIVE_TARGETS): Remove.
2322 * Makefile.in: Regenerate.
2323 * aclocal.m4: Regenerate.
2324 * doc/Makefile.am (install, install-info): Remove.
2325 (install-data-local): A new hook for install-info.
2326 (AUTOMAKE_OPTIONS): Require automake 1.8.
2327 * doc/Makefile.in: Regenerate.
2329 2004-01-02 Nutan Singh <nutan@kpitcummins.com>
2331 * doc/c-sh.texi: Update description about floating point behavior
2334 2004-01-02 Bernardo Innocenti <bernie@develer.com>
2336 * configure.in: Add m68k-uClinux target.
2337 * configure: Regenerate.
2339 For older changes see ChangeLog-0203
2345 version-control: never
projects / external / binutils.git / blob